Sam's Notebook

Resazurin Assays - USDA M.gigas Families in Response to Temperature Stress

Sam White — Wed, 15 Apr 2026 07:00:00 GMT

INTRO

Quick notebook post to review the resazurin assay data from the USDA M.gigas families in response to temperature stress. This is a work in progress, but I wanted to share some preliminary findings and visualizations. Notebook will be updated later with more details on the methods and results.

RESULTS

Plots

Normalized Fluorescence Over Time - Line Plots

Mean normalized fluorescence by family over time.

Mean normalized fluorescence over time for Family 1.

Mean normalized fluorescence over time for Family 2.

Mean normalized fluorescence over time for Family 3.

Mean normalized fluorescence over time for Family 5.

Mean normalized fluorescence over time for Family 6.

Mean normalized fluorescence over time for Family 7.

Mean normalized fluorescence over time for Family 8.

Mean normalized fluorescence over time for Family 9.

Mean normalized fluorescence over time for Family 10.

Normalized Fluorescence Over Time - Box Plots

Normalized fluorescence by family.

Normalized fluorescence for Family 1.

Normalized fluorescence for Family 2.

Normalized fluorescence for Family 3.

Normalized fluorescence for Family 5.

Normalized fluorescence for Family 6.

Normalized fluorescence for Family 7.

Normalized fluorescence for Family 8.

Normalized fluorescence for Family 9.

Normalized fluorescence for Family 10.

Delta Fluorescence (Raw) Over Time

Delta mean fluorescence by family over time (raw).

Delta fluorescence over time for Family 1 (raw).

Delta fluorescence over time for Family 2 (raw).

Delta fluorescence over time for Family 3 (raw).

Delta fluorescence over time for Family 5 (raw).

Delta fluorescence over time for Family 6 (raw).

Delta fluorescence over time for Family 7 (raw).

Delta fluorescence over time for Family 8 (raw).

Delta fluorescence over time for Family 9 (raw).

Delta fluorescence over time for Family 10 (raw).

Samples Submitted - Geoduck Gonad Histology

Sam White — Tue, 14 Apr 2026 07:00:00 GMT

INTRO

Submitted 10 geoduck gonad samples for histological processing. These samples were collected in lab on 20260401 by Steven and people from Washington Fish & Wildlife. Gonad tissue was sample, placed in histology cassettes, and fixed in 10% formalin in filtered seawater for 24hrs. After fixation, samples were transferred to 70% ethanol for storage until processing.

Cassettes were submitted to the University of Washington Pathology Research Services Laboratory in Health Sciences Building C411 for processing and histological sectioning. The lab will embed the samples in paraffin, section them, and stain with hematoxylin and eosin (H&E) for microscopic examination. The goal of this histological analysis is to assess the reproductive status and gonadal development of the geoducks, which will provide insights into their reproductive biology and potential impacts of environmental factors on their reproduction.

Here’s the list of samples submitted:

4-M-IN
3-M-X
4-A-X
2-A-X
1-A-X
2-P-IN
4-P-IN
3-M-IN
1-P-IN
3-P-IN

Sample submission on lab bench.

Genome Comparisons - Lake Trout PacBio Assemblies Flye vs hifiasm Using QUAST

Sam White — Fri, 10 Apr 2026 07:00:00 GMT

INTRO

I previously assembled the Lake Trout genomes for the lean and siscowet ecotypes using both Flye and hifiasm with PacBio data (GitHub Issue). I wanted to compare the two assemblies to see which one was better. I used QUAST to compare the assemblies for some basic stats.

Notebook links to assemblies:

Below is the rendered markdown from 13.2-genome-assembly-comparisons.Rmd.

1 BACKGROUND

This will use QUAST (Mikheenko et al. 2018, 2023) to evaluate the two sets of PacBio genome assemblies produced with Flye(GitHub) and hifiasm (GitHub) for both ecotypes.

2 SETUP

2.1 Libraries

library(knitr)
library(tidyverse)

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.5.2     ✔ tibble    3.2.1
## ✔ lubridate 1.9.4     ✔ tidyr     1.3.1
## ✔ purrr     1.0.4     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package () to force all conflicts to become errors

library(reticulate)

knitr::opts_chunk$set(
  echo = TRUE,         # Display code chunks
    eval = FALSE,        # Evaluate code chunks
    results = "hold",   # Hold outputs and show them after the full code chunk
  warning = FALSE,     # Hide warnings
        collapse = FALSE,    # Keep code and output in separate blocks
        warning = FALSE,     # Hide warnings
        message = FALSE,     # Hide messages
        comment = "##"      # Prefix output lines with '##' so output is visually distinct
)

2.2 Set R variables

# OUTPUT DIRECTORY
output_dir <- "../output/13.2-genome-assembly-comparisons"

# INPUT FILES
flye_lean_fasta <- "../output/13.1-genome-assembly-lean/snam-lean-pb-flye-assembly.fasta"
flye_siscowet_fasta <- "../output/13.1-genome-assembly-siscowet/snam-siscowet-pb-flye-assembly.fasta"
hifiasm_lean_fasta <- "../output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-assembly.fa"
hifiasm_siscowet_fasta <- "../output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-assembly.fa"

# SETTINGS
threads <- "40"


# Export these as environment variables for bash chunks.
Sys.setenv(
  flye_lean_fasta = flye_lean_fasta,
  flye_siscowet_fasta = flye_siscowet_fasta,
  hifiasm_lean_fasta = hifiasm_lean_fasta,
  hifiasm_siscowet_fasta = hifiasm_siscowet_fasta,
  output_dir = output_dir,
  threads = threads
)

3 Run QUAST

# Make output directory, if it doesn't exist
mkdir --parents "${output_dir}"

quast.py \
"${flye_lean_fasta}" \
"${hifiasm_lean_fasta}" \
"${flye_siscowet_fasta}" \
"${hifiasm_siscowet_fasta}" \
--output-dir "${output_dir}" \
--threads "${threads}" \
--labels "lean_flye, lean_hifiasm, siscowet_flye, siscowet_hifiasm"

## /srlab/programs/miniforge3-24.7.1-0/bin/quast.py:4: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
##   __import__('pkg_resources').run_script('quast==5.3.0', 'quast.py')
## /srlab/programs/miniforge3-24.7.1-0/lib/python3.12/site-packages/quast-5.3.0-py3.12.egg/EGG-INFO/scripts/quast.py ../output/13.1-genome-assembly-lean/snam-lean-pb-flye-assembly.fasta ../output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-assembly.fa ../output/13.1-genome-assembly-siscowet/snam-siscowet-pb-flye-assembly.fasta ../output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-assembly.fa --output-dir ../output/13.2-genome-assembly-comparisons --threads 40 --labels lean_flye, lean_hifiasm, siscowet_flye, siscowet_hifiasm
## 
## Version: 5.3.0
## 
## System information:
##   OS: Linux-4.18.0-513.18.1.el8_9.x86_64-x86_64-with-glibc2.39 (linux_64)
##   Python version: 3.12.5
##   CPUs number: 192
## 
## Started: 2026-04-10 12:49:08
## 
## Logging to /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.2-genome-assembly-comparisons/quast.log
## 
## CWD: /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/code
## Main parameters: 
##   MODE: default, threads: 40, min contig length: 500, min alignment length: 65, min alignment IDY: 95.0, \
##   ambiguity: one, min local misassembly length: 200, min extensive misassembly length: 1000
## 
## WARNING: Can't draw plots: python-matplotlib is missing or corrupted.
## 
## Contigs:
##   Pre-processing...
##   1  ../output/13.1-genome-assembly-lean/snam-lean-pb-flye-assembly.fasta ==> lean_flye
##   2  ../output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-assembly.fa ==> lean_hifiasm
##   3  ../output/13.1-genome-assembly-siscowet/snam-siscowet-pb-flye-assembly.fasta ==> siscowet_flye
##   4  ../output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-assembly.fa ==> siscowet_hifiasm
## 
## 2026-04-10 12:51:06
## Running Basic statistics processor...
##   Contig files: 
##     1  lean_flye
##     2  lean_hifiasm
##     3  siscowet_flye
##     4  siscowet_hifiasm
##   Calculating N50 and L50...
##     1  lean_flye, N50 = 74514, L50 = 10897, auN = 172736.7, Total length = 4046355081, GC % = 42.89, # N's per 100 kbp =  0.00
##     2  lean_hifiasm, N50 = 325701, L50 = 2542, auN = 608980.5, Total length = 3733314870, GC % = 42.57, # N's per 100 kbp =  0.00
##     3  siscowet_flye, N50 = 78564, L50 = 10934, auN = 189833.9, Total length = 4336770215, GC % = 43.17, # N's per 100 kbp =  0.00
##     4  siscowet_hifiasm, N50 = 514031, L50 = 1844, auN = 874464.1, Total length = 4008351139, GC % = 42.84, # N's per 100 kbp =  0.00
## Done.
## 
## NOTICE: Genes are not predicted by default. Use --gene-finding or --glimmer option to enable it.
## 
## 2026-04-10 12:56:51
## Creating large visual summaries...
## This may take a while: press Ctrl-C to skip this step..
##   1 of 1: Creating Icarus viewers...
## Done
## 
## 2026-04-10 12:58:19
## RESULTS:
##   Text versions of total report are saved to /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.2-genome-assembly-comparisons/report.txt, report.tsv, and report.tex
##   Text versions of transposed total report are saved to /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.2-genome-assembly-comparisons/transposed_report.txt, transposed_report.tsv, and transposed_report.tex
##   HTML version (interactive tables and plots) is saved to /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.2-genome-assembly-comparisons/report.html
##   Icarus (contig browser) is saved to /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.2-genome-assembly-comparisons/icarus.html
##   Log is saved to /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.2-genome-assembly-comparisons/quast.log
## 
## Finished: 2026-04-10 12:58:19
## Elapsed time: 0:09:10.519647
## NOTICEs: 1; WARNINGs: 1; non-fatal ERRORs: 0
## 
## Thank you for using QUAST!

RESULTS

All output files are here:

https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.2-genome-assembly-comparisons/

QUAST Report (HTML):

13.2-genome-assembly-comparisons/report.html

Assembly	lean_flye	lean_hifiasm	siscowet_flye	siscowet_hifiasm
# contigs (>= 0 bp)	112562	33035	110158	24184
# contigs (>= 1000 bp)	112160	33035	109812	24184
# contigs (>= 5000 bp)	106095	33002	104705	24177
# contigs (>= 10000 bp)	81886	31353	84765	23609
# contigs (>= 25000 bp)	37973	21447	41932	17566
# contigs (>= 50000 bp)	17083	13735	18607	12107
Total length (>= 0 bp)	4046369763	3733314870	4336778941	4008351139
Total length (>= 1000 bp)	4046083421	3733314870	4336539268	4008351139
Total length (>= 5000 bp)	4023676535	3733170075	4318011635	4008320038
Total length (>= 10000 bp)	3840245419	3719588210	4166152008	4003479664
Total length (>= 25000 bp)	3119711632	3550261715	3452967778	3897289116
Total length (>= 50000 bp)	2400080225	3275969678	2647693108	3704640383
# contigs	112527	33035	110135	24184
Largest contig	2170611	5276269	2753427	8967193
Total length	4046355081	3733314870	4336770215	4008351139
GC (%)	42.89	42.57	43.17	42.84
N50	74514	325701	78564	514031
N90	14295	41727	15962	65625
auN	172736.7	608980.5	189833.9	874464.1
L50	10897	2542	10934	1844
L90	65455	15577	64363	10408
# N’s per 100 kbp	0	0	0	0

QUAST Report

QUAST Genome Size Plots

DISCUSSION

In general, it appears that the hifiasm assemblies are better than the flye assemblies, with higher N50 and larger contigs. However, the Flye assemblies have a higher total length (~7% more).

REFERENCES

Mikheenko, Alla, Andrey Prjibelski, Vladislav Saveliev, Dmitry Antipov, and Alexey Gurevich. 2018. “Versatile Genome Assembly Evaluation with QUAST-LG.” Bioinformatics 34 (13): i142–50. https://doi.org/10.1093/bioinformatics/bty266.

Mikheenko, Alla, Vladislav Saveliev, Pascal Hirsch, and Alexey Gurevich. 2023. “WebQUAST: Online Evaluation of Genome Assemblies.” Nucleic Acids Research 51 (W1): W601–6. https://doi.org/10.1093/nar/gkad406.

Resazurin Assays - Selected Juvenile Ruditapes philippinarum Exposed to 36C Acute Heat Stress Day 1

Sam White — Wed, 01 Apr 2026 07:00:00 GMT

INTRO

This is the second day of a two day experiment. Day 0 corresponds to the first day of measurements, and Day 1 corresponds to the second day of measurements. The same clams were measured on both days. Day 0 notebook entry is available here:

Day 0 Notebook Entry

MATERIALS & METHODS

Control clams (C) compared with selected clams (S) at 36^oC.

Clams were processed in 12-well plates and submerged in 5.0mL of resazurin working solution prepared today.

986.66 mL filtered seawater (4^oC filtered sea water from 11/19/2025 by AH)
2.22 mL resazurin stock solution (from step 1 above)
1.00 mL DMSO
10.00 mL antibiotic solution (100x Penn/Strep & 100x Fungizone)

Plates were measured every 30mins on a Synergy HTX (Agilent) plate reader over the course of 4hrs.

Plate layout was randomized; Steven is currently the only person who knows the treatment assignments.

Important

Low volumes were noted in the following plates/wells, which may have been due to clam spitting:

DA1 BA3 DB2 EC2

Plate temps

Wells were spot checked and range noted in table below.

PLATE	TIME	TEMP
A	0	11-12
A	1	29
A	2	31
A	3	32-33
A	4	33-34

RESULTS

Data are available here:

https://github.com/RobertsLab/resazurin-assay-development/tree/main/data/clam/20260401-clam-Day2-36C-C-vs-S

At the time of this writing, there is no plate layout file available for this experiment, so it’s not possible to determine which wells correspond to which clams. Steven is working on creating a plate layout file, which will be added to the GitHub repository when it’s available.

Resazurin Assays - Selected Juvenile Ruditapes philippinarum Exposed to 36C Acute Heat Stress

Sam White — Wed, 01 Apr 2026 07:00:00 GMT

INTRO

Steven asked me to perform resazurin assays on “selected” juvenile Ruditapes philippinarum (Manila clam) exposed to 36^oC acute heat stress. The goal of the experiment is to compare metabolic activity between the “selected” clams and control clams when exposed to heat stress. This is similar to the previous experiment at 36^oC Day 0 Notebook Entry and Day 1 Notebook Entry, but with a set of larger clams.

MATERIALS & METHODS

Clams were processed in 12-well plates and submerged in 5.0mL of resazurin working solution stored at 4^oC,prepared on 20260330 (notebook entry).

Plates were measured every 30mins on a Synergy HTX (Agilent) plate reader.

Plate layout was randomized; Steven is currently the only person who knows the treatment assignments.

Important

The following wells lost volume - possibly due to clam spitting?

Plate Temps

Wells were spot checked and range noted in table below.

PLATE	TIME	TEMP
G	0	11-12
G	0.5	14-16
G	1	24-30
G	1.5	27-30
G	2	31-32
G	2.5	31-32
G	3	31-32
G	3.5	31-32
G	4	33

RESULTS

Data are available here:

https://github.com/RobertsLab/resazurin-assay-development/tree/main/data/clam/20260401-clam-36C-C-vs-S

Genome Assembly - S.namaycush Lean Ecotype with PacBio HiFi Reads Using hifiasm on Hyak

Sam White — Tue, 31 Mar 2026 07:00:00 GMT

INTRO

As part of project-lake-trout (GitHub repo), Steven asked that I assemble the genome of a lean ecotype of lake trout (GitHub Issue), Salvelinus namaycush, using PacBio HiFi reads. I previously assembled this genome using Flye, on 20230326 (notebook entry), but Steven asked me to also try hifiasm (GitHub Issue)for the assembly.

I performed the assembly on Hyak (Klone), UW’s high-performance computing cluster. I used our Apptainer (Singularity) container to run the job (srlab-R4.4-bioinformatics-container-e5bcfea.sif), which includes hifiasm version 0.25.0.

Below is the rendered markdown from 13.1-hifiasm-genome-assembly-lean.Rmd. The run was very fast; it only took a day and half, vs the ~7 days for the Flye assembly.

PRIMARY OUTPUTS

hifiasm produces three assemblies: a primary assembly, and two haplotype-resolved assemblies. Below are the links to the FastA files. The outputs are only GFA graph files. I had to convert them to FastA using gfatools.

Primary assembly:

FastA (3.5GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-assembly.fa
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-assembly.fa.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-assembly.fa.fai

Haplotype 1:

FastA (3.2GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-hap1-assembly.fa
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-hap1-assembly.fa.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-hap1-assembly.fa.fai

Haplotype 2:

FastA (2.7GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-hap2-assembly.fa
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-hap2-assembly.fa.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/pb-hifiasm-lean-hap2-assembly.fa.fai

After this, I’ll compare the two assemblies (Flye vs. hifiasm) using QUAST, which will provide a more comprehensive set of assembly statistics and metrics.

1 BACKGROUND

Use hifiasm (GitHub) (Cheng et al. 2021, 2022, 2024) to assemble PacBio reads for S.namaycush lean ecotype.

Due to large file sizes, the outputs will not be on GitHub. They may be found here instead:

https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-lean/

2 SETUP

2.1 Libraries

2.2 Set R variables

# OUTPUT DIRECTORY
data_dir <- "/mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/data/pacbio-lean"
output_dir <- "../output/13.1-hifiasm-genome-assembly-lean"

# PROGRAMS
gfatools <- c("/srlab/programs/gfatools/gfatools")
hifiasm <- c("/srlab/programs/miniforge3-24.7.1-0/envs/hifiasm-0.25.0_env/bin/hifiasm")


# SETTINGS
threads <- "40"


# Export these as environment variables for bash chunks.
Sys.setenv(
  data_dir = data_dir,
  gfatools = gfatools,
  hifiasm=hifiasm,
  output_dir = output_dir,
  threads = threads
)

# Make output directory, if it doesn't exist
mkdir --parents "${output_dir}"

cd "${output_dir}"

# Create an array of .fastq.gz files
fastqs=(${data_dir}/*.fastq.gz)

# Print the result (optional, for verification)
## newline-delimited format
echo "List of FastQs used for assembly:"

printf "%s\n" "${fastqs[@]}" \
| tee input_fastqs.txt

# Run hifiasm assembly
"${hifiasm}" \
-o pb-hifiasm-lean-assembly.asm \
-t "${threads}" \
"${fastqs[@]}" \
> pb-hifiasm-lean-assembly.log 2>&1

3 GFATOOLS

3.1 Primary contigs to FastA

This will extract a FastA file from the graph files output by hifiasm.

cd "${output_dir}"

"${gfatools}" gfa2fa \
pb-hifiasm-lean-assembly.asm.bp.p_ctg.gfa \
> pb-hifiasm-lean-assembly.fa

samtools faidx pb-hifiasm-lean-assembly.fa

3.2 Haplotype 1 contigs to FastA

This will extract a FastA file from the graph files output by hifiasm.

cd "${output_dir}"

"${gfatools}" gfa2fa \
pb-hifiasm-lean-assembly.asm.bp.hap1.p_ctg.gfa \
> pb-hifiasm-lean-hap1-assembly.fa

samtools faidx pb-hifiasm-lean-hap1-assembly.fa

3.3 Haplotype 2 contigs to FastA

This will extract a FastA file from the graph files output by hifiasm.

cd "${output_dir}"

"${gfatools}" gfa2fa \
pb-hifiasm-lean-assembly.asm.bp.hap2.p_ctg.gfa \
> pb-hifiasm-lean-hap2-assembly.fa

samtools faidx pb-hifiasm-lean-hap2-assembly.fa

3.4 Generate checksums

cd "${output_dir}"

for fasta in *.fa
do
  md5sum "${fasta}" \
  | tee "${fasta}".md5
done

## 4ba0c3cc7a7ab090d67047cb22f10bb2  pb-hifiasm-lean-assembly.fa
## 77d48fe452af73d751039245f1bd688e  pb-hifiasm-lean-hap1-assembly.fa
## 21b9cfeb83ec8ce522fcc624873b58eb  pb-hifiasm-lean-hap2-assembly.fa

4 REFERENCES

Cheng, Haoyu, Mobin Asri, Julian Lucas, Sergey Koren, and Heng Li. 2024. “Scalable Telomere-to-Telomere Assembly for Diploid and Polyploid Genomes with Double Graph.” Nature Methods 21 (6): 967–70. https://doi.org/10.1038/s41592-024-02269-8.

Cheng, Haoyu, Gregory T. Concepcion, Xiaowen Feng, Haowen Zhang, and Heng Li. 2021. “Haplotype-Resolved de Novo Assembly Using Phased Assembly Graphs with Hifiasm.” Nature Methods 18 (2): 170–75. https://doi.org/10.1038/s41592-020-01056-5.

Cheng, Haoyu, Erich D. Jarvis, Olivier Fedrigo, et al. 2022. “Haplotype-Resolved Assembly of Diploid Genomes Without Parental Data.” Nature Biotechnology 40 (9): 1332–35. https://doi.org/10.1038/s41587-022-01261-x.

Genome Assembly - S.namaycush Siscowet Ecotype with PacBio HiFi Reads Using hifiasm on Hyak

Sam White — Tue, 31 Mar 2026 07:00:00 GMT

INTRO

As part of project-lake-trout (GitHub repo), Steven asked that I assemble the genome of a siscowet ecotype of lake trout (GitHub Issue), Salvelinus namaycush, using PacBio HiFi reads. I previously assembled this genome using Flye, on 20260326 (notebook entry), but Steven asked me to also try hifiasm (GitHub Issue)for the assembly.

Below is the rendered markdown from 13.1-hifiasm-genome-assembly-siscowet.Rmd. The run was very fast; it only took a day and half.

PRIMARY OUTPUTS

Primary assembly:

FastA (3.8GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-assembly.fa
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-assembly.fa.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-assembly.fa.fai

Haplotype 1:

FastA (3.4GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-hap1-assembly.fa
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-hap1-assembly.fa.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-hap1-assembly.fa.fai

Haplotype 2:

FastA (3.2GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-hap2-assembly.fa
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-hap2-assembly.fa.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/pb-hifiasm-siscowet-hap2-assembly.fa.fai

After this, I’ll compare the two assemblies (Flye vs. hifiasm) using QUAST, which will provide a more comprehensive set of assembly statistics and metrics.

1 BACKGROUND

Use hifiasm (GitHub) (Cheng et al. 2021, 2022, 2024) to assemble PacBio reads for S.namaycush siscowet ecotype.

Due to large file sizes, the outputs will not be on GitHub. They may be found here instead:

https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-hifiasm-genome-assembly-siscowet/

2 SETUP

2.1 Libraries

2.2 Set R variables

# OUTPUT DIRECTORY
data_dir <- "/mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/data/pacbio-siscowet"
output_dir <- "../output/13.1-hifiasm-genome-assembly-siscowet"

# PROGRAMS
gfatools <- c("/srlab/programs/gfatools/gfatools")
hifiasm <- c("/srlab/programs/miniforge3-24.7.1-0/envs/hifiasm-0.25.0_env/bin/hifiasm")


# SETTINGS
threads <- "40"


# Export these as environment variables for bash chunks.
Sys.setenv(
  data_dir = data_dir,
  gfatools = gfatools,
  hifiasm=hifiasm,
  output_dir = output_dir,
  threads = threads
)

3 RUN HIFIASM

# Make output directory, if it doesn't exist
mkdir --parents "${output_dir}"

cd "${output_dir}"

# Create an array of .fastq.gz files
fastqs=(${data_dir}/*.fastq.gz)

# Print the result (optional, for verification)
## newline-delimited format
echo "List of FastQs used for assembly:"

printf "%s\n" "${fastqs[@]}" \
| tee input_fastqs.txt

# Run hifiasm assembly
"${hifiasm}" \
-o pb-hifiasm-siscowet-assembly.asm \
-t "${threads}" \
"${fastqs[@]}" \
> pb-hifiasm-siscowet-assembly.log 2>&1

4 GFATOOLS

4.1 Primary contigs to FastA

This will extract a FastA file from the graph files output by hifiasm.

cd "${output_dir}"

"${gfatools}" gfa2fa \
pb-hifiasm-siscowet-assembly.asm.bp.p_ctg.gfa \
> pb-hifiasm-siscowet-assembly.fa

samtools faidx pb-hifiasm-siscowet-assembly.fa

4.2 Haplotype 1 contigs to FastA

This will extract a FastA file from the graph files output by hifiasm.

cd "${output_dir}"

"${gfatools}" gfa2fa \
pb-hifiasm-siscowet-assembly.asm.bp.hap1.p_ctg.gfa \
> pb-hifiasm-siscowet-hap1-assembly.fa

samtools faidx pb-hifiasm-siscowet-hap1-assembly.fa

4.3 Haplotype 2 contigs to FastA

This will extract a FastA file from the graph files output by hifiasm.

cd "${output_dir}"

"${gfatools}" gfa2fa \
pb-hifiasm-siscowet-assembly.asm.bp.hap2.p_ctg.gfa \
> pb-hifiasm-siscowet-hap2-assembly.fa

samtools faidx pb-hifiasm-siscowet-hap2-assembly.fa

4.4 Generate checksums

cd "${output_dir}"

for fasta in *.fa
do
  md5sum "${fasta}" \
  | tee "${fasta}".md5
done

## d41d8cd98f00b204e9800998ecf8427e  pb-hifiasm-lean-assembly.fa
## c5e88b1d81585383b3fb7354aa624f34  pb-hifiasm-siscowet-assembly.fa
## cb85945e4b07c42fba548865e3039d00  pb-hifiasm-siscowet-hap1-assembly.fa
## 5d813ba82405afef60628e3c8325306f  pb-hifiasm-siscowet-hap2-assembly.fa

5 REFERENCES

Resazurin Assays - Selected Juvenile Ruditapes philippinarum Exposed to 36C Acute Heat Stress Day 0

Sam White — Tue, 31 Mar 2026 07:00:00 GMT

INTRO

This is the first day of a two day experiment. Day 0 corresponds to the first day of measurements, and Day 1 corresponds to the second day of measurements. The same clams were measured on both days. Day 1 notebook entry is available here:

Day 1 Notebook Entry

MATERIALS & METHODS

Control clams (C) compared with selected clams (S) at 36^oC.

Clams were processed in 12-well plates and submerged in 5.0mL of resazurin working solution stored at 4^oC,prepared on 20260330 (notebook entry).

Plates were measured every 30mins on a Synergy HTX (Agilent) plate reader.

Plate layout was randomized; Steven is currently the only person who knows the treatment assignments.

Important

Seemed to be inconsistent heating, with interior wells usually ~1.5^oC lower than rest of wells.

The following wells lost volume - possibly due to clam spitting?

Plate A, C3
Plate C, C4
Plate B, A3
Plate D, B2

After 3.5hrs, Steven removed the resazurin working solution from the wells and replaced it with 5.0mL of seawater. Plates were returned to the cold room and will be measured again tomorrow.

RESULTS

Data are available here:

https://github.com/RobertsLab/resazurin-assay-development/tree/main/data/clam/20260331-clam-36C-C-vs-S

Resazurin Assays - Selected Juvenile Ruditapes philippinarum Exposed to 40C Acute Heat Stress

Sam White — Mon, 30 Mar 2026 07:00:00 GMT

INTRO

Steven asked me to perform resazurin assays on “selected” juvenile Ruditapes philippinarum (Manila clam) exposed to 40^oC acute heat stress. The goal of the experiment is to determine if the “selected” clams have higher metabolic activity than control clams when exposed to heat stress.

Data and results are available here:

https://github.com/RobertsLab/resazurin-assay-development/tree/main/data/clam/20260330-clam-40C-C-vs-S

MATERIALS & METHODS

Steven picked 48 clams from the “Selected” group and 48 clams from the “Control” group.

Two plates per treatment were measured: “blue” and “red”. The colors reflect the coloration of the clams.

Blanks were included on each plate, and consisted of wells with resazurin working solution, but no clams. Blanks were in Column 6 of each plate.

Plates were kept in a small, table-top, fan-powered incubator at 40^oC.

Resazurin Working Solution

Clams were processed in 24-well plates and submerged in 2.2mL of resazurin working solution; prepared by Sam White 20260330:

986.66 mL filtered seawater (4^oC filtered sea water from 11/19/2025 by AH)
2.22 mL resazurin stock solution (from step 1 above)
1.00 mL DMSO
10.00 mL antibiotic solution (100x Penn/Strep & 100x Fungizone)

Measurements

Plates were measured every 30mins on a Synergy HTX (Agilent) plate reader. The plate reader was set to measure fluorescence with an excitation wavelength of 530nm and an emission wavelength of 590nm.

RESULTS

Here are line plots of raw fluorescence, by well, across all time points. Black lines correspond to the Blanks.

![Line plots of raw fluorescence, by well, across all time points. Black lines correspond to the Blanks.](https://github.com/RobertsLab/resazurin-assay-development/blob/main/data/clam/20260330-clam-40C-C-vs-S/plots/fluorescence-by-well_all-plates.png?raw=true()

Note

The S-red plate shows changes in fluorescence in the Blanks, suggesting contamination, or, more likely, accidental splashing/wicking of neighboring wells. The volume in the wells was extremely high, so it’s possible that the resazurin working solution was splashed/wicked into the Blank wells via the lid, which had visible resazurin dropletsduring handling.

The S-blue plate shows the overall highest levels of raw fluorescence. Since these plots do not take into account the size of the clams, it’s difficult to draw many conclusions from these data.

Additionally, since the “Control” and “Selected” clams were not randomly assigned to the “red” and “blue” plates, it’s possible that the differences in fluorescence between the plates are due to plate effects, rather than treatment effects.

Samples Submitted - Salmon Kidney Tissue to SolusCell for Cristian Gallardo

Sam White — Mon, 30 Mar 2026 07:00:00 GMT

INTRO

Steven asked me to submit some salmon kidney tissue samples to SolusCell for RNA-seq (GitHub Issue). The samples are for Cristian Gallardo from the University of Concepción in Chile. Samples were shipped on dry ice via FedEx on 2026-03-30. The tracking number is: 870130241319.

Genome Assembly - S.namaycush Lean Ecotype with PacBio HiFi Reads Using Flye

Sam White — Thu, 26 Mar 2026 07:00:00 GMT

INTRO

As part of project-lake-trout (GitHub repo), Steven asked that I assemble the genome of a lean ecotype of lake trout (GitHub Issue), Salvelinus namaycush, using PacBio HiFi reads. I opted to use Flye for the assembly, as we’d previously used it for Chionoecetes bairdi (Tanner crab) (Wikipedia), and it performed well. The assembly process was straightforward and quick.

Below is the rendered markdown from 13.1-genome-assembly-lean.Rmd. The run took ~7 days.

PRIMARY OUTPUTS

FastA (3.9GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-genome-assembly-lean/snam-lean-pb-flye-assembly.fasta
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-genome-assembly-lean/snam-lean-pb-flye-assembly.fasta.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-genome-assembly-lean/snam-lean-pb-flye-assembly.fasta.fai

Once I’ve also assembled this genome using hifiasm, I’ll compare the two assemblies using QUAST, which will provide a more comprehensive set of assembly statistics and metrics.

13.1-genome-assembly-lean

Sam White 2026-03-26

1 BACKGROUND

Use Flye(GitHub) (Kolmogorov et al. 2019) to assemble PacBio reads for S.namaycush lean ecotype.

2 SETUP

2.1 Libraries

library(knitr)
library(tidyverse)

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.5.2     ✔ tibble    3.2.1
## ✔ lubridate 1.9.4     ✔ tidyr     1.3.1
## ✔ purrr     1.0.4     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package () to force all conflicts to become errors

library(reticulate)

knitr::opts_chunk$set(
  echo = TRUE,         # Display code chunks
    eval = FALSE,        # Evaluate code chunks
    results = "hold",   # Hold outputs and show them after the full code chunk
  warning = FALSE,     # Hide warnings
        collapse = FALSE,    # Keep code and output in separate blocks
        warning = FALSE,     # Hide warnings
        message = FALSE,     # Hide messages
        comment = "##"      # Prefix output lines with '##' so output is visually distinct
)

2.2 Set R variables

# OUTPUT DIRECTORY
data_dir <- "../data/pacbio-lean"
output_dir <- "../output/13.1-genome-assembly-lean"

#OUTPUT FILE(S)
prefix <- "snam-lean-pb-flye"


# CONDA
conda_env_name <- c("/srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env")
conda_path <- c("/srlab/programs/miniforge3-24.7.1-0/bin/conda")

# SETTINGS
genome_size <- "2.3g"
threads <- "32"


# Export these as environment variables for bash chunks.
Sys.setenv(
  data_dir = data_dir,
  genome_size = genome_size,
  output_dir = output_dir,
  prefix = prefix,
  threads = threads
)

2.3 Load Flye conda environment

If this is successful, the first line of output should show that the Python being used is the one in your Flye conda environment path.

E.g.

python: /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env/bin/python

use_condaenv(condaenv = conda_env_name, conda = conda_path)
py_config()

## python:         /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env/bin/python
## libpython:      /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env/lib/libpython3.12.so
## pythonhome:     /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env:/srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env
## version:        3.12.13 | packaged by conda-forge | (main, Mar  5 2026, 16:50:00) [GCC 14.3.0]
## numpy:           [NOT FOUND]
## 
## NOTE: Python version was forced by use_python() function

# Make output directory, if it doesn't exist
mkdir --parents "${output_dir}"

# Create an array of .fastq.gz files
fastqs=(${data_dir}/*.fastq.gz)

# Print the result (optional, for verification)
## newline-delimited format
echo "List of FastQs used for assembly:"

printf "%s\n" "${fastqs[@]}" \
| tee "${output_dir}/input_fastqs.txt"

# Run Flye assembly
flye \
--pacbio-hifi "${fastqs[@]}" \
--genome-size "${genome_size}" \
--out-dir "${output_dir}" \
--threads "${threads}" \
> "${output_dir}"/pb-lean-assembly.log 2>&1

3 OUTPUTS

3.1 List files

ls -ltrh "${output_dir}"

## total 8.2G
## -rw-r--r-- 1 samwhite all  326 Mar 26 11:02 input_fastqs.txt
## drwxr-xr-x 2 samwhite all  512 Mar 30 07:44 00-assembly
## drwxr-xr-x 2 samwhite all 8.0K Mar 30 12:12 10-consensus
## drwxr-xr-x 2 samwhite all  512 Mar 31 08:22 20-repeat
## drwxr-xr-x 2 samwhite all  512 Mar 31 08:33 30-contigger
## drwxr-xr-x 2 samwhite all 8.0K Mar 31 17:25 40-polishing
## -rw-r--r-- 1 samwhite all   92 Mar 31 17:25 params.json
## -rw-r--r-- 1 samwhite all  53M Mar 31 17:25 assembly_graph.gv
## -rw-r--r-- 1 samwhite all 4.0G Mar 31 17:25 assembly_graph.gfa
## -rw-r--r-- 1 samwhite all 3.9G Mar 31 17:25 assembly.fasta
## -rw-r--r-- 1 samwhite all 4.8M Mar 31 17:25 assembly_info.txt
## -rw-r--r-- 1 samwhite all 3.2K Mar 31 17:25 pb-lean-assembly.log
## -rw-r--r-- 1 samwhite all 338M Mar 31 17:25 flye.log
## -rw-r--r-- 1 samwhite all 3.9M Apr  2 10:05 assembly.fasta.fai

3.2 Rename files

Files have generic “assembly_*” naming.

Renaming to reflect specific species-ecotype-data-assembly_method.

cd "${output_dir}"

for file in assembly*
do
  mv "${file}" ${prefix}-${file}
done

3.3 Create FastA Index

cd "${output_dir}"

samtools faidx ${prefix}-assembly.fasta

3.4 Create checksums

cd "${output_dir}"

for file in ${prefix}*
do
  md5sum "${file}" \
  | tee "${file}".md5
done

## ac7c2af6c32834f86954cfb9893c4c50  snam-lean-pb-flye-assembly.fasta
## 242ea029692207d1c7df0b7e842fd58f  snam-lean-pb-flye-assembly.fasta.fai
## 35f1973b873fd7498ebeeecdb7eecd2e  snam-lean-pb-flye-assembly_graph.gfa
## d4f6a3a41846f6285d0290c7a41ed1f7  snam-lean-pb-flye-assembly_graph.gv
## e1224b9a628253fd8cee1d39517343ef  snam-lean-pb-flye-assembly_info.txt

3.5 Assembly stats

cd "${output_dir}"

tail flye.log

## [2026-03-31 17:25:45] root: INFO: Assembly statistics:
## 
##  Total length:   4046369763
##  Fragments:  112562
##  Fragments N50:  74514
##  Largest frg:    2170611
##  Scaffolds:  0
##  Mean coverage:  15
## 
## [2026-03-31 17:25:45] root: INFO: Final assembly: /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.1-genome-assembly-lean/assembly.fasta

Kolmogorov, Mikhail, Jeffrey Yuan, Yu Lin, and Pavel A. Pevzner. 2019. “Assembly of Long, Error-Prone Reads Using Repeat Graphs.” Nature Biotechnology 37 (5): 540–46. https://doi.org/10.1038/s41587-019-0072-8.

Genome Assembly - S.namaycush Siscowet Ecotype with PacBio HiFi Reads Using Flye

Sam White — Thu, 26 Mar 2026 07:00:00 GMT

INTRO

As part of project-lake-trout (GitHub repo), Steven asked that I assemble the genome of a siscowet ecotype of lake trout (GitHub Issue), Salvelinus namaycush, using PacBio HiFi reads. I opted to use Flye for the assembly, as we’d previously used it for Chionoecetes bairdi (Tanner crab) (Wikipedia), and it performed well. The assembly process was straightforward and quick.

Below is the rendered markdown from 13.1-genome-assembly-siscowet.Rmd. The run took ~7 days.

PRIMARY OUTPUTS

FastA (4.1GB): https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-genome-assembly-siscowet/snam-siscowet-pb-flye-assembly.fasta
MD5: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-genome-assembly-siscowet/snam-siscowet-pb-flye-assembly.fasta.md5
FastA index: https://gannet.fish.washington.edu/gitrepos/project-lake-trout/output/13.1-genome-assembly-siscowet/snam-siscowet-pb-flye-assembly.fasta.fai

Once I’ve also assembled this genome using hifiasm, I’ll compare the two assemblies using QUAST, which will provide a more comprehensive set of assembly statistics and metrics.

1 BACKGROUND

Use Flye(GitHub) (Kolmogorov et al. 2019) to assemble PacBio reads for S.namaycush siscowet ecotype.

2 SETUP

2.1 Libraries

2.2 Set R variables

# OUTPUT DIRECTORY
data_dir <- "../data/pacbio-siscowet"
output_dir <- "../output/13.1-genome-assembly-siscowet"

#OUTPUT FILE(S)
prefix <- "snam-siscowet-pb-flye"



# CONDA
conda_env_name <- c("/srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env")
conda_path <- c("/srlab/programs/miniforge3-24.7.1-0/bin/conda")

# SETTINGS
genome_size <- "2.3g"
threads <- "100"


# Export these as environment variables for bash chunks.
Sys.setenv(
  data_dir = data_dir,
  genome_size = genome_size,
  output_dir = output_dir,
  prefix = prefix,
  threads = threads
)

2.3 Load Flye conda environment

If this is successful, the first line of output should show that the Python being used is the one in your Flye conda environment path.

E.g.

python: /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env/bin/python

use_condaenv(condaenv = conda_env_name, conda = conda_path)
py_config()

## python:         /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env/bin/python
## libpython:      /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env/lib/libpython3.12.so
## pythonhome:     /srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env:/srlab/programs/miniforge3-24.7.1-0/envs/flye-2.9.6-env
## version:        3.12.13 | packaged by conda-forge | (main, Mar  5 2026, 16:50:00) [GCC 14.3.0]
## numpy:           [NOT FOUND]
## 
## NOTE: Python version was forced by use_python() function

3 FLYE

# Make output directory, if it doesn't exist
mkdir --parents "${output_dir}"

# Create an array of .fastq.gz files
fastqs=(${data_dir}/*.fastq.gz)

# Print the result (optional, for verification)
## newline-delimited format
echo "List of FastQs used for assembly:"

printf "%s\n" "${fastqs[@]}" \
| tee "${output_dir}/input_fastqs.txt"

# Run Flye assembly
flye \
--pacbio-hifi "${fastqs[@]}" \
--genome-size "${genome_size}" \
--out-dir "${output_dir}" \
--threads "${threads}" \
> "${output_dir}"/pb-siscowet-assembly.log 2>&1

4 OUTPUTS

4.1 List files

ls -ltrh "${output_dir}"

## total 8.9G
## -rw-r--r-- 1 samwhite all  342 Apr  6 08:08 input_fastqs.txt
## drwxr-xr-x 2 samwhite all  512 Apr  8 11:51 00-assembly
## drwxr-xr-x 2 samwhite all 8.0K Apr  8 15:08 10-consensus
## drwxr-xr-x 2 samwhite all  512 Apr  9 02:23 20-repeat
## drwxr-xr-x 2 samwhite all  512 Apr  9 02:40 30-contigger
## drwxr-xr-x 2 samwhite all 8.0K Apr  9 10:29 40-polishing
## -rw-r--r-- 1 samwhite all   92 Apr  9 10:29 params.json
## -rw-r--r-- 1 samwhite all  55M Apr  9 10:29 assembly_graph.gv
## -rw-r--r-- 1 samwhite all 4.3G Apr  9 10:30 assembly_graph.gfa
## -rw-r--r-- 1 samwhite all 4.2G Apr  9 10:30 assembly.fasta
## -rw-r--r-- 1 samwhite all 4.8M Apr  9 10:30 assembly_info.txt
## -rw-r--r-- 1 samwhite all 3.2K Apr  9 10:30 pb-siscowet-assembly.log
## -rw-r--r-- 1 samwhite all 482M Apr  9 10:30 flye.log

4.2 Rename files

Files have generic “assembly_*” naming.

Renaming to reflect specific species-ecotype-data-assembly_method.

cd "${output_dir}"

for file in assembly*
do
  mv "${file}" ${prefix}-${file}
done

4.3 Create FastA Index

cd "${output_dir}"

samtools faidx ${prefix}-assembly.fasta

4.4 Create checksums

cd "${output_dir}"

for file in ${prefix}*
do
  md5sum "${file}" \
  | tee "${file}".md5
done

## 30139d7d8d7fdad878543fc5a5f3f0a8  snam-siscowet-pb-flye-assembly.fasta
## adf9dfc0996f12c28a41089e148e693a  snam-siscowet-pb-flye-assembly.fasta.fai
## 62f9467fdd049e7dafaf7c63871c4184  snam-siscowet-pb-flye-assembly_graph.gfa
## 949d7663ab4d54500e00856fd80deed9  snam-siscowet-pb-flye-assembly_graph.gv
## 1f336e131d47f582e15bdd0acc66414f  snam-siscowet-pb-flye-assembly_info.txt

4.5 Assembly stats

cd "${output_dir}"

tail flye.log

## [2026-04-09 10:30:36] root: INFO: Assembly statistics:
## 
##  Total length:   4336778941
##  Fragments:  110158
##  Fragments N50:  78564
##  Largest frg:    2753427
##  Scaffolds:  0
##  Mean coverage:  21
## 
## [2026-04-09 10:30:36] root: INFO: Final assembly: /mmfs1/gscratch/scrubbed/samwhite/gitrepos/RobertsLab/project-lake-trout/output/13.1-genome-assembly-siscowet/assembly.fasta

Software Installation - EGAPx and FCS-GX on Klone

Sam White — Thu, 05 Mar 2026 08:00:00 GMT

INTRO

Laura asked me to install EGAPx and FCS-GX on Klone (GitHub Issue) so that she can use them for annotation of some existing snow crab RNA-seq data she is working with.

METHODS

[FCS-GX]

The steps below follow those outlined in the FCS-GX documentation.

Get FCS Python script

cd /gscratch/srlab/containers \
&& curl -LO https://github.com/ncbi/fcs/raw/main/dist/fcs.py

Get FCS-GX Singularity image

curl https://ftp.ncbi.nlm.nih.gov/genomes/TOOLS/FCS/releases/latest/fcs-gx.sif -Lo fcs-gx.sif

Get FCS-GX databases

This downloads ~500GB of data, so be sure you have enough storage space before running this command! Additionally, the download may take a while to complete.

module load coenv/python/3.13.11

cd /gscratch/scrubbed/samwhite/databases

SOURCE_DB_MANIFEST="https://ncbi-fcs-gx.s3.amazonaws.com/gxdb/latest/all.manifest"

python3 /gscratch/srlab/containers/fcs.py \
--image /gscratch/srlab/containers/fcs-gx.sif \
db \
get --mft "$SOURCE_DB_MANIFEST" \
--dir "./gxdb"

Screenshot of completed DB downloads

Installation was not tested by running an annotation job, but the above steps were successfully completed without issue.

EGAPx

The steps below follow those outlined in the EGAPx documentation.

Containers and scripts

I created an Apptainer (Singularity) container for EGAPx and a script to execute it.

Container definition file: srlab-NCBI-EGAPx.def (GitHub)

This container is based on the EGAPx Dockerfile and includes EGAPx and all dependencies.
Primarily written using Claude Sonnet 4.6 (gAI agent) with some manual edits to ensure it would build correctly.

Execution script: run_egapx.sh (GitHub).

This script is designed to execute the srlab-NCBI-EGAPx.sif container. It will create a cache directory, set a cache variable, and execute the container.

Both have been added to Klone and can be found in the /mmfs1/gscratch/srlab/containers directory.

Container and script have been added to Klone:

/mmfs1/gscratch/srlab/containers/run_egapx.sh: Script to execute srlab-NCBI-EGAPx.sif container. This will create a cache directory, set a cache variable, and bind directories to the container.

Example usage: bash ./run_egapx.sh \ /opt/egapx/examples/input_D_farinae_small.yaml \ -e singularity \ -w ./work \ -o ./output

This requires that the srlab-NCBI-EGAPx.sif container is in the same directory as this script. Both can be copied to other directories, if desired.
/mmfs1/gscratch/srlab/containers/srlab-NCBI-EGAPx.sif: Apptainer (Singularity) container. Can be executed by itself, but is better to use with the run_egapx.sh to ensure cache directory is set.

To see additional help examples for using the srlab-NCBI-EGAPx.sif container:

singularity run-help srlab-NCBI-EGAPx.sif

Test EGAPx container

The built-in test example was used to test the container. The test completed successfully, with all expected output files generated.

Screenshot of EGAPx test output

CONCLUSION

EGAPx and FCS-GX have been successfully installed on Klone. Both were tested to ensure they are working correctly, with FCS-GX tested by downloading the databases and EGAPx tested by running the built-in test example.

For info on how to run EGAPx: https://github.com/ncbi/egapx

Software Testing - Stevens workflow-annotation Pipeline

Sam White — Thu, 05 Mar 2026 08:00:00 GMT

INTRO

Steven asked me to test out his workflow-annotation pipeline (GitHub) after Hannah encountered an issue yesterday and we implemented a potential fix (GitHub Issue).

The pipeline was designed to take nucleotide sequences all the way through gene ontology annotation.

After some back-and-forth attempts at troubleshooting, I resorted to asking the Gemini 3 Flash (preview) gAI agent. The suggestion was the lack of an User-Agent in the URL request header was causing the 403 permission denied error Hannah (and I) encountered.

I implemented the suggested changes and tested them.

METHODS

As input, I used a subset of a FastA file provided by Hannah (see code below).

Set up Conda environment

Create Mamba/Conda environment

Important

Be sure to deactivate any current Python/Conda environments!

deactivate

and/or

conda deactivate

mamba create \
-n sr320-workflow-annotation \
python=3.11 \
diamond \
blast \
requests \
pandas \
goatools \
matplotlib \
-c bioconda \
-c conda-forge \
-y

R setup

Code

De Novo Transcriptome Assembly and Annotation - E5 A.pulchra RNA-seq Using Trinity and PASA

Sam White — Fri, 27 Feb 2026 08:00:00 GMT

INTRO

This notebook performs a comprehensive transcriptome assembly and annotation for A.pulchra using the PASA (Program to Assemble Spliced Alignments), as part of the E5 timeseries_molecular project. This produced an updated genome annotation that incorporates the RNA-seq data, including alternative splicing information. However, it’s important to note that the resulting GFF/BED files will not contain any annotations from the original genome GFF which did not have any support from RNA-seq alignments!

Thus, the resulting PASA annotations will be a subset of the original genome annotations, but with updated gene models based on the RNA-seq data. The PASA pipeline will merge the de novo and genome-guided transcriptome assemblies, clean the transcripts, and update the genome annotations with alternative splicing information where supported by the data.

Due to large file sizes, the majority of important output files are not availabe on GitHub, but can be accessed on Gannet here:

https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/D-Apul/output/00.30-D-Apul-transcriptome-assembly-Trinity/

Primary products:

Trinity de-novo assembly FastA: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/D-Apul/output/00.30-D-Apul-transcriptome-assembly-Trinity/de_novo_assembly/apul-denovo-Trinity.fasta
Trinity genome-guided assembly FastA: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/D-Apul/output/00.30-D-Apul-transcriptome-assembly-Trinity/genome_guided_assembly/apul-GG-Trinity.fasta
PASA final GFF: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/D-Apul/output/00.30-D-Apul-transcriptome-assembly-Trinity/PASA/apul-PASA.gff3
PASA final BED: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/D-Apul/output/00.30-D-Apul-transcriptome-assembly-Trinity/PASA/apul-PASA.bed

The markdown below was produced from the original Rmd script:

00.30-D-Apul-transcriptome-assembly-Trinity.Rmd (GitHub)

Not described below is the access to the PASA web portal, which allows the user to browse the PASA results in a user-friendly interface. This can be accessed here:

http://gannet.fish.washington.edu:9000/

Enter apul_pasa as the database name. If this is your first time using the database, it will take many minutes for the data to load (the MySQL files behind the scenes are ~10GB in size, so it takes a while to process). After initial acces, the data is cached and access is much faster.

Screenshot of apul_pasa web portal on Gannet

1 BACKGROUND

This notebook performs a comprehensive transcriptome assembly and annotaiton for A.pulchra using the PASA (Program to Assemble Spliced Alignments) pipeline, as well as alternative isoform identification. This pipeline relies on both de novo and genome-guided transcriptome assemblies with Trinity.

The key steps include: 1. De Novo Assembly and Genome-Guided Assembly: Using Trinity to assemble transcripts directly from RNA-seq reads. 2. PASA Pipeline: Utilizing PASA to merge the assemblies, clean transcripts, and update genome annotations with alternative splicing information.

Programs Used: - Trinity - Transcriptome assembly. - PASA - Annotation update and transcript assembly refinement. - AGAT - GFF/GTF toolkit for annotation merging and conversion. - Singularity (Apptainer) - Container platform used to run assembly pipelines.

1.1 Expected outputs

TRINITY:

FastA: De novo transcriptome assembly.
FastA: Genome-guided transcriptome assembly.

PASA:

GFF3: Genome annotations produced by PASA pipeline in GFF3.
BED: Genome annotations produced by PASA pipeline in BED format.

2 SETUP

2.1 Libraries and markdown settings

library(knitr)
knitr::opts_chunk$set(
  echo = TRUE,         # Display code chunks
  eval = FALSE,        # Evaluate code chunks
  warning = FALSE,     # Hide warnings
  message = FALSE,     # Hide messages
  comment = ""         # Prevents appending '##' to beginning of lines in code output
)

2.2 Set variables

# DIRECTORIES
top_output_dir <- file.path("..", "output")

output_dir <- file.path(top_output_dir, "00.30-D-Apul-transcriptome-assembly-Trinity")
de_novo_output_dir <- file.path(output_dir, "de_novo_assembly")
genome_guided_output_dir <- file.path(output_dir, "genome_guided_assembly")
pasa_container_dir <- file.path("/home", "shared", "containers")
PASA_HOME <- "/usr/local/src/PASApipeline"
pasa_output_dir <- file.path(output_dir, "PASA")
stringtie_gtf_dir <- file.path(top_output_dir, "02.20-D-Apul-RNAseq-alignment-HiSat2")
trimmed_reads_dir <- file.path(top_output_dir, "01.00-D-Apul-RNAseq-trimming-fastp-FastQC-MultiQC")


# FILES
bam_alignment <- file.path(top_output_dir, "02.20-D-Apul-RNAseq-alignment-HiSat2", "sorted-bams-merged.bam")

## Path for genome will be relative to PASA output dir
genome_fasta <- file.path("..", "..", "..", "data", "Apulchra-genome.fa")
genome_gff <- file.path("..", "data", "Apulchra-genome.gff")
denovo_assembly_name <- "apul-denovo-Trinity"
genome_guided_assembly_name <- "apul-GG-Trinity"
pasa_bed <- "apul-PASA.bed"
pasa_container <- "pasapipeline.v2.5.3.simg"
pasa_gff <- "apul-PASA.gff3"
stringtie_gtf <- file.path(stringtie_gtf_dir, "Apulchra-genome.stringtie.gtf")

#SETTINGS
## THREADS
threads <- "44"

## MAX RAM
max_ram <- "100G"

# PROGRAMS
samtools <- file.path("/home", "shared", "samtools-1.12", "samtools")



# FORMATTING
line <- "-----------------------------------------------"

# Export these as environment variables for bash chunks.
Sys.setenv(
  bam_alignment = bam_alignment,
  denovo_assembly_name = denovo_assembly_name,
  de_novo_output_dir = de_novo_output_dir,
  genome_fasta = genome_fasta,
  genome_gff = genome_gff,
  genome_guided_assembly_name = genome_guided_assembly_name,
  genome_guided_output_dir = genome_guided_output_dir,
  line = line,
  max_ram = max_ram,
  output_dir = output_dir,
  top_output_dir = top_output_dir,
  pasa_container = pasa_container,
  pasa_container_dir = pasa_container_dir,
  pasa_bed = pasa_bed,
  pasa_gff = pasa_gff,
  PASA_HOME = PASA_HOME,
  pasa_output_dir = pasa_output_dir,
  samtools = samtools,
  stringtie_gtf_dir = stringtie_gtf_dir,
  stringtie_gtf = stringtie_gtf,
  threads = threads,
  trimmed_reads_dir = trimmed_reads_dir
)

3 DE NOVO ASSEMBLY

3.1 Run Trinity

Trinity was run using the Trinity Singularity (Apptainer) container, trinityrnaseq.v2.15.2.simg from the urol-e5/timeseries_molecular/D-Apul/code/ directory.

This was done in a terminal, outside of this notebook.

3.1.1 Set Bash variables

# Directories
top_output_dir="../output"

output_dir="${top_output_dir}/00.30-D-Apul-transcriptome-assembly-Trinity"
de_novo_output_dir="${output_dir}/de_novo_assembly"
genome_guided_output_dir="${output_dir}/genome_guided_assembly"
trimmed_reads_dir="${top_output_dir}/01.00-D-Apul-RNAseq-trimming-fastp-FastQC-MultiQC"

# PASA INPUT FILES
####### NEED TO BE RELATIVE TO PASA SUBDIRECTORY #######
genome_fasta="../../../data/Apulchra-genome.fa"
genome_gff="../../../data/Apulchra-genome.gff3"
pasa_container="pasapipeline.v2.5.3.simg"
PASA_HOME="/usr/local/src/PASApipeline"
stringtie_gtf="../../02.20-D-Apul-RNAseq-alignment-HiSat2/Apulchra-genome.stringtie.gtf"

## THREADS
threads="44"

## MAX RAM
max_ram="100G"

# Make output directoy, if it doesn't exist
mkdir --parents ${de_novo_output_dir}

## Inititalize arrays
R1_array=()
R2_array=()

# Variables for R1/R2 lists
R1_list=""
R2_list=""

# Create array of fastq R1 files
R1_array=(${trimmed_reads_dir}/*R1_001.fastp-trim.fq.gz)

# Create array of fastq R2 files
R2_array=(${trimmed_reads_dir}/*R2_001.fastp-trim.fq.gz)

# Create list of fastq files used in analysis
## Uses parameter substitution to strip leading path from filename
if [ ! -f "${de_novo_output_dir}/fastq.list.txt" ]; then
  for fastq in ${trimmed_reads_dir}/*.fq.gz
  do
    echo "${fastq##*/}" >> ${de_novo_output_dir}/fastq.list.txt
  done
fi

# Create comma-separated lists of FastQ reads
R1_list=$(echo "${R1_array[@]}" | tr " " ",")
R2_list=$(echo "${R2_array[@]}" | tr " " ",")

3.1.2 Run Trinity Singularity image.

Used “stranded” setting (–SS_lib_type).

singularity exec \
-B /home \
-e trinityrnaseq.v2.15.2.simg \
Trinity \
--seqType fq \
--max_memory ${max_ram} \
--CPU ${threads} \
--SS_lib_type RF \
--left "${R1_list}" \
--right "${R2_list}" \
--output ${de_novo_output_dir}/trinity_out_dir \
--full_cleanup \
> ${de_novo_output_dir}/trinity.log \
2>&1

3.2 Rename output files

3.2.1 Rename FastA

# Rename generic assembly FastA
mv ${de_novo_output_dir}/trinity_out_dir.Trinity.fasta \
${de_novo_output_dir}/${denovo_assembly_name}.fasta

3.2.2 Rename gene map and log

mv ${de_novo_output_dir}/trinity_out_dir.Trinity.fasta.gene_trans_map \
${de_novo_output_dir}/${denovo_assembly_name}.gene_trans_map

mv ${de_novo_output_dir}/trinity.log \
${de_novo_output_dir}/${denovo_assembly_name}.log

3.2.3 Assembly stats

3.2.3.1 Run Trinity Singularity image.

singularity exec -B /home \
-e trinityrnaseq.v2.15.2.simg \
/usr/local/bin/util/TrinityStats.pl \
../output/00.30-D-Apul-transcriptome-assembly-Trinity/de_novo_assembly/apul-denovo-Trinity.fasta \
> ../output/00.30-D-Apul-transcriptome-assembly-Trinity/de_novo_assembly/apul-denovo-Trinity.stats

3.3 Create FastA index

${samtools} faidx \
${de_novo_output_dir}/${denovo_assembly_name}.fasta

3.4 Checksums

cd ${de_novo_output_dir}

md5sum ${denovo_assembly_name}.fasta | tee ${denovo_assembly_name}.fasta.md5

f5a8213c889f1fdbd8a48b5047e8d797  apul-denovo-Trinity.fasta

4 GENOME-GUIDED ASSEMBLY

Trinity was run using the Trinity Singularity (Apptainer) container, trinityrnaseq.v2.15.2.simg from the urol-e5/timeseries_molecular/D-Apul/code/ directory.

This was done in a terminal, outside of this notebook.

singularity exec \
-B /home -e trinityrnaseq.v2.15.2.simg \
Trinity \
--genome_guided_bam ../output/02.20-D-Apul-RNAseq-alignment-HiSat2/sorted-bams-merged.bam \
--genome_guided_max_intron 10000 \
--max_memory ${max_ram} \
--CPU ${threads} \
--SS_lib_type RF \
--output ${genome_guided_output_dir}/trinity_out_dir \
--full_cleanup \
> ${genome_guided_output_dir}/trinity.log 2>&1

4.1 Rename output files

4.1.1 Rename FastA

# Rename generic assembly FastA
mv ${genome_guided_output_dir}/trinity_out_dir.Trinity-GG.fasta \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta

4.1.2 Rename gene map and log

mv ${genome_guided_output_dir}/trinity_out_dir.Trinity-GG.fasta.gene_trans_map \
${genome_guided_output_dir}/${genome_guided_assembly_name}.gene_trans_map

mv ${genome_guided_output_dir}/trinity.log \
${genome_guided_output_dir}/${genome_guided_assembly_name}.log

4.2 Create FastA index

${samtools} faidx \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta

4.3 Checksums

cd ${genome_guided_output_dir}

md5sum ${genome_guided_assembly_name}.fasta | tee ${genome_guided_assembly_name}.fasta.md5

44c4f7b6493239377cf02d9f9d5fb15f  apul-GG-Trinity.fasta

4.3.1 Assembly stats

4.3.1.1 Run Trinity Singularity image.

singularity exec -B /home \
-e trinityrnaseq.v2.15.2.simg \
/usr/local/bin/util/TrinityStats.pl \
../output/00.30-D-Apul-transcriptome-assembly-Trinity/genome_guided_assembly/apul-GG-Trinity.fasta \
> ../output/00.30-D-Apul-transcriptome-assembly-Trinity/genome_guided_assembly/apul-GG-Trinity.stats

5 PASA PIPELINE

5.1 Concatenate Trinity assemblies

cat ${de_novo_output_dir}/${denovo_assembly_name}.fasta \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta \
> ${pasa_output_dir}/transcripts.fasta

5.1.1 Confirm counts

# Count transcripts in each file
denovo_count=$(grep -c "^>" ${de_novo_output_dir}/${denovo_assembly_name}.fasta)
genome_guided_count=$(grep -c "^>" ${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta)
pasa_count=$(grep -c "^>" ${pasa_output_dir}/transcripts.fasta)

# Calculate sum of first two counts
sum=$((denovo_count + genome_guided_count))

# Compare sum to PASA count
echo "De novo count: $denovo_count"
echo "Genome-guided count: $genome_guided_count"
echo "Sum: $sum"
echo "PASA count: $pasa_count"

if [ $sum -eq $pasa_count ]; then
    echo "✓ Counts match: $sum = $pasa_count"
else
    echo "✗ Counts do not match: $sum ≠ $pasa_count (difference: $((pasa_count - sum)))"
fi

De novo count: 1854167
Genome-guided count: 661982
Sum: 2516149
PASA count: 2516149
✓ Counts match: 2516149 = 2516149

5.2 Extract transcript accessions

singularity exec \
-B /home \
-e ${pasa_container_dir}/${pasa_container} \
$PASA_HOME/misc_utilities/accession_extractor.pl \
< ${de_novo_output_dir}/${denovo_assembly_name}.fasta \
> ${pasa_output_dir}/tdn.accs

head ${pasa_output_dir}/tdn.accs

5.3 Clean transcripts

cd ${pasa_output_dir}

singularity exec \
-B /home \
-e \
--env USER="$USER" \
${pasa_container} \
$PASA_HOME/bin/seqclean \
transcripts.fasta \
-c 16

5.4 PASA Assembly

5.4.1 Fix schema key length

cd ${pasa_output_dir}

#### Fix schema key length issue ####
singularity exec ${pasa_container} \
cat $PASA_HOME/schema/cdna_alignment_mysqlschema \
> cdna_alignment_mysqlschema

# Fix all variations of gene_id and model_id indexes
sed -i 's/KEY gene_id_idx (gene_id)/KEY gene_id_idx (gene_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/KEY mod_idx (model_id)/KEY mod_idx (model_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/(gene_id)/(gene_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/(model_id)/(model_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/KEY gene_idx (annotation_version,gene_id)/KEY gene_idx (annotation_version,gene_id(255))/g' cdna_alignment_mysqlschema

5.4.2 Run PASA Assembly Pipeline

This was executed outside of RStudio due to the verbose output, which will cause RStudio to crash.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
--config alignAssembly.config \
--create \
--run \
--genome ${genome_fasta} \
--transcripts transcripts.fasta.clean \
--trans_gtf ${stringtie_gtf} \
--ALT_SPLICE \
-T \
-u transcripts.fasta \
--ALIGNERS blat,gmap,minimap2 \
--TDN tdn.accs \
--transcribed_is_aligned_orient \
--annot_compare \
-L \
--annots ${genome_gff} \
--TRANSDECODER \
--CPU ${threads}

5.4.3 Alternative Splicing

This doesn’t seem to have run during the assembly phase, so ran separately.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
-c alignAssembly.config \
--ALT_SPLICE \
-g ${genome_fasta} \
-t all.transcripts.fasta.clean \
--CPU ${threads}

5.4.4 Update annotations

Now includes alternative splicing info.

Uses output GFF3 from initial annotations as annotation input.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
-c annotCompare.config \
--annot_compare \
-L \
--annots apul_pasa.gene_structures_post_PASA_updates.2550130.gff3 \
-g ${genome_fasta} \
-t all.transcripts.fasta.clean \
--CPU ${threads}

6 PASA OUTPUTS

6.1 Generate checksums

cd "${pasa_output_dir}"

md5sum apul_pasa.gene_structures_post_PASA_updates.3761114.gff3 | tee apul_pasa.gene_structures_post_PASA_updates.3761114.gff3.md5

md5sum apul_pasa.gene_structures_post_PASA_updates.3761114.bed | tee apul_pasa.gene_structures_post_PASA_updates.3761114.bed.md5

66809687062caaf68e2a5bf118a77398  apul_pasa.gene_structures_post_PASA_updates.3761114.gff3
341d5aa495756bafcc5753722e8c93d4  apul_pasa.gene_structures_post_PASA_updates.3761114.bed

6.2 Rename outputs

cd "${pasa_output_dir}"
cp apul_pasa.gene_structures_post_PASA_updates.3761114.gff3 "${pasa_gff}"

cp apul_pasa.gene_structures_post_PASA_updates.3761114.bed "${pasa_bed}"

md5sum "${pasa_gff}" | tee "${pasa_gff}".md5
md5sum "${pasa_bed}" | tee "${pasa_bed}".md5

66809687062caaf68e2a5bf118a77398  apul-PASA.gff3
341d5aa495756bafcc5753722e8c93d4  apul-PASA.bed

6.3 GFF3 Preview

head -n 50 "${pasa_output_dir}"/"${pasa_gff}"

# ORIGINAL: FUN_044110-T1 original gene structure, not modified by PASA
ptg000092l  funannotate gene    6557    14524   .   +   .   ID=FUN_044110;Name=FUN_044110-T1
ptg000092l  funannotate mRNA    6557    14524   .   +   .   ID=FUN_044110-T1;Parent=FUN_044110;Name=FUN_044110-T1
ptg000092l  funannotate exon    6557    6575    .   +   .   ID=FUN_044110-T1.exon1;Parent=FUN_044110-T1
ptg000092l  funannotate CDS 6557    6575    .   +   0   ID=FUN_044110-T1.cds.1;Parent=FUN_044110-T1
ptg000092l  funannotate exon    8536    8657    .   +   .   ID=FUN_044110-T1.exon2;Parent=FUN_044110-T1
ptg000092l  funannotate CDS 8536    8657    .   +   2   ID=FUN_044110-T1.cds.2;Parent=FUN_044110-T1
ptg000092l  funannotate exon    9330    9410    .   +   .   ID=FUN_044110-T1.exon3;Parent=FUN_044110-T1
ptg000092l  funannotate CDS 9330    9410    .   +   0   ID=FUN_044110-T1.cds.3;Parent=FUN_044110-T1
ptg000092l  funannotate exon    10230   10307   .   +   .   ID=FUN_044110-T1.exon4;Parent=FUN_044110-T1
ptg000092l  funannotate CDS 10230   10307   .   +   0   ID=FUN_044110-T1.cds.4;Parent=FUN_044110-T1
ptg000092l  funannotate exon    12512   14524   .   +   .   ID=FUN_044110-T1.exon5;Parent=FUN_044110-T1
ptg000092l  funannotate CDS 12512   14524   .   +   0   ID=FUN_044110-T1.cds.5;Parent=FUN_044110-T1


#PROT FUN_044110-T1 FUN_044110  MLIFNRGELLFLPIRNVLCMMPQGYKNALPGYKDLYLSQAITEEVHNMFSTGIDCGVSSFKHGPSLSLLALDKKLCVIALIESLFQVRGLPFCARSLNKTALSTQSSLMSPHYLPLQSNQLGQVKPTFQLSEPGRSDENFSDTDPKFICKHPRIHVPTSVGVVQPSVRRRTSDNPVSPTESRSESPLFSLLHEPETSVATTTLGDPTNQLVSRTLAKSNVNQLSAQDIPNDSSVQQNSLESHLTPLNQPVDPDPLLPESIDGTSSIEIDSSKESTKSSRTVTLSESEMSPQLRLDLEEIRKFYSLPINLNRDGGVLQDVSIGKMLERIKGFLWFLKKVKGVEPALTYCINPEVLQQFVEFMMKNRGIKAITCSRYVTSLISACKVPLACTQDEQKEESLEKIRAIQRQLERLSRQEKIDSDSLNPQTDKVVYSELLELCREFKWEVSEKTGADRARSCMNLCLLLMYCAVNPGRVKEYISLRIYKDQSGDQLKDQNFIWFKEDGGIVLLENNYKTRNTYGLNTTDVSSVTYLNYYLQLYKSKMRSLLLHGNDHDFFFVAPRGNRFSHASYNYYISGLFEKYLSRRLTTVDLRKIVVNYFLSLPESGDYSLRESFATLMKHSIRAQQKYYDERPLTQKKDRALDLLTSVARRSLDEDEPEIVSDEDQEGYLDCLPVPGDFVALVAANSTEKVPEVFVAKVLRLSEDKKTAYLADFAEEEPGRFKSKAGKSYKENTNSLIFPIDIVFSHSDGLYELRTPKIDLHLVTVQKKS*

# ORIGINAL: FUN_044109-T1 original gene structure, not modified by PASA
ptg000092l  funannotate gene    3791    5008    .   +   .   ID=FUN_044109;Name=FUN_044109-T1
ptg000092l  funannotate mRNA    3791    5008    .   +   .   ID=FUN_044109-T1;Parent=FUN_044109;Name=FUN_044109-T1
ptg000092l  funannotate exon    3791    3811    .   +   .   ID=FUN_044109-T1.exon1;Parent=FUN_044109-T1
ptg000092l  funannotate CDS 3791    3811    .   +   0   ID=FUN_044109-T1.cds.1;Parent=FUN_044109-T1
ptg000092l  funannotate exon    4856    5008    .   +   .   ID=FUN_044109-T1.exon2;Parent=FUN_044109-T1
ptg000092l  funannotate CDS 4856    5008    .   +   0   ID=FUN_044109-T1.cds.2;Parent=FUN_044109-T1


#PROT FUN_044109-T1 FUN_044109  MSNFSIKSEDEPASPLVRTDKNEELPTIPQAFRKLKECYCNCSWGDSLHADPQLTGE*

# ORIGINAL: FUN_044111-T1 original gene structure, not modified by PASA
ptg000092l  funannotate gene    17683   22752   .   +   .   ID=FUN_044111;Name=FUN_044111-T1
ptg000092l  funannotate mRNA    17683   22752   .   +   .   ID=FUN_044111-T1;Parent=FUN_044111;Name=FUN_044111-T1
ptg000092l  funannotate exon    17683   17881   .   +   .   ID=FUN_044111-T1.exon1;Parent=FUN_044111-T1
ptg000092l  funannotate CDS 17683   17881   .   +   0   ID=FUN_044111-T1.cds.1;Parent=FUN_044111-T1
ptg000092l  funannotate exon    18082   18144   .   +   .   ID=FUN_044111-T1.exon2;Parent=FUN_044111-T1
ptg000092l  funannotate CDS 18082   18144   .   +   2   ID=FUN_044111-T1.cds.2;Parent=FUN_044111-T1
ptg000092l  funannotate exon    18985   19058   .   +   .   ID=FUN_044111-T1.exon3;Parent=FUN_044111-T1
ptg000092l  funannotate CDS 18985   19058   .   +   2   ID=FUN_044111-T1.cds.3;Parent=FUN_044111-T1
ptg000092l  funannotate exon    19708   19968   .   +   .   ID=FUN_044111-T1.exon4;Parent=FUN_044111-T1
ptg000092l  funannotate CDS 19708   19968   .   +   0   ID=FUN_044111-T1.cds.4;Parent=FUN_044111-T1
ptg000092l  funannotate exon    20540   21849   .   +   .   ID=FUN_044111-T1.exon5;Parent=FUN_044111-T1
ptg000092l  funannotate CDS 20540   21849   .   +   0   ID=FUN_044111-T1.cds.5;Parent=FUN_044111-T1
ptg000092l  funannotate exon    22188   22752   .   +   .   ID=FUN_044111-T1.exon6;Parent=FUN_044111-T1
ptg000092l  funannotate CDS 22188   22752   .   +   1   ID=FUN_044111-T1.cds.6;Parent=FUN_044111-T1


#PROT FUN_044111-T1 FUN_044111  MAFVRLPRPLLDEFITLWDQVQSISPTLPEQARHLIHRVDDKVAEIRSQASDTPTTGGHGAESQSTAADQPPQIPCLAMQEETASSSDDQPATPLVRKRKRRVAQKITSSAQGIQLPARPPDIVSEPTPDINFESMEIKYLKQALASVVQPSTKAEVYLKYLAKCDFNTTLEPHSIIKFNQDDVRRMVGIGQNPNGEEEVVGKLHQALQQRLQLAVQQAFSIGEFMSTCIKEHGSTLPDEDHQRRRGRPRVSSLLDTLDSIESLGASSSCLQDQVLIFQAISNFPLLKYVNEPMTHFVCNDVRLAIRHLPAALHHKCDVPENIQDAHINLRPFRSLVSDQESPPLSGDMSYSNVTPCDGDGLVVGINLDHREPLENAEYTRFYGIDAPELSSVHFIKTNDFQHVFCKQVGHISLCAVHLFLQMFLLSGSAKLCEELPREAAPQPRDIYNRALKEYWFKFITPPSQHLEKVFLQSLEELVPPTSESRKRLMSPFPATMATAANPFLLSLNALLVVSGFCHVFTKYCQDGFLLGLQAIARDNKLGPIWCGASRKFIFGCTSGNNTDFFLKHFTPETTSHLARAGFPFKHSNAFLPWHERQMLKQLCSQETTRTAARNHLAQHLPGMEPQFGMYIDIQRSNQGEGYQTVRTGEAYLKVMNSQVVGIGGNSTGLGLFTLKKIPRGTLVCAYAPTATIWEGKLNGDYVLETSFNNKVISVNGKENLFELGLGIYCNDGSFPFSLARARFSNVISHRVNCEYCKCGDGIWLKTVRDVSAGEELLMCYSQDGSYWATIFSREQLNQITAALNSCGPSLQDAERCIRLLQV*

# PASA_UPDATE: FUN_044120-T1, single gene model update, valid-1, status:[pasa:asmbl_497397,status:12], valid-1
# PASA_UPDATE: FUN_044120-T1.1.69943284, single gene model update, valid-1, status:[pasa:asmbl_497398,status:12], valid-1
# PASA_UPDATE: FUN_044120-T1.2.69943284, single gene model update, valid-1, status:[pasa:asmbl_497396,status:12], valid-1

6.4 GFF Comparisons

printf '%s\n\n' "Original GFF feature counts:"
awk '!/^#/ && !/^[[:space:]]*$/ && NF > 0 && $3 != "" {print $3}' ${genome_gff} \
| sort | uniq -c | sort -rn | awk '{print $2, $1}'

echo ""
echo "${line}"
echo ""

printf "%s\n\n" "Updated GFF feature counts:"
awk -F "\t" '!/^#/ && !/^[[:space:]]*$/ && NF > 0 && $3 != "" {print $3}' "${pasa_output_dir}"/"${pasa_gff}" \
| sort | uniq -c | sort -rn | awk '{print $2, $1}'

Original GFF feature counts:

exon 209537
CDS 201613
gene 44371
mRNA 36447
tRNA 7924

-----------------------------------------------

Updated GFF feature counts:

exon 273865
CDS 268202
mRNA 42845
gene 36742
five_prime_UTR 13538
three_prime_UTR 13463

7 EXTRACT PROTEINS TO FASTA

cd "${pasa_output_dir}"

awk '/^#PROT / {print ">" $2 "." $3 "\n" $4}' "${pasa_gff}" > proteins-PASA.fasta

printf "%s\n\n" "Original protein counts:"
grep --count "^#PROT" "${pasa_gff}"

echo ""
echo "${line}"
echo ""

printf "%s\n\n" "Extracted protein counts:"
grep --count "^>" proteins-PASA.fasta

# Create FastA Index
${samtools} faidx proteins-PASA.fasta

Original protein counts:

42845

-----------------------------------------------

Extracted protein counts:

42845

7.1 Checksums

cd "${pasa_output_dir}"
md5sum proteins-PASA.fasta | tee proteins-PASA.fasta.md5

b8bcef2f9bc2e19edf908343c07c4448  proteins-PASA.fasta

De Novo Transcriptome Assembly and Annotation - E5 P.evermanni RNA-seq Using Trinity and PASA

Sam White — Fri, 27 Feb 2026 08:00:00 GMT

INTRO

This notebook performs a comprehensive transcriptome assembly and annotation for P.evermanni using the PASA (Program to Assemble Spliced Alignments), as part of the E5 timeseries_molecular project. This produced an updated genome annotation that incorporates the RNA-seq data, including alternative splicing information. However, it’s important to note that the resulting GFF/BED files will not contain any annotations from the original genome GFF which did not have any support from RNA-seq alignments!

Due to large file sizes, the majority of important output files are not availabe on GitHub, but can be accessed on Gannet here:

https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/

Primary products:

Trinity de-novo assembly FastA: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/de_novo_assembly/peve-denovo-Trinity.fasta
Trinity genome-guided assembly FastA: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/genome_guided_assembly/peve-GG-Trinity.fasta
PASA final GFF: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/PASA/peve-PASA.gff3
PASA final BED: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/PASA/peve-PASA.bed

The markdown below was produced from the original Rmd script:

00.30-E-Peve-transcriptome-assembly-Trinity.Rmd (GitHub)

Not described below is the access to the PASA web portal, which allows the user to browse the PASA results in a user-friendly interface. This can be accessed here:

http://gannet.fish.washington.edu:9000/

Enter peve_pasa as the database name. If this is your first time using the database, it will take many minutes for the data to load (the MySQL files behind the scenes are ~10GB in size, so it takes a while to process). After initial acces, the data is cached and access is much faster.

Screenshot of peve_pasa web portal on Gannet

1 BACKGROUND

This notebook performs a comprehensive transcriptome assembly and annotaiton for P.evermanni using the PASA (Program to Assemble Spliced Alignments) pipeline, as well as alternative isoform identification. This pipeline relies on both de novo and genome-guided transcriptome assemblies with Trinity.

1.1 Expected outputs

TRINITY:

FastA: De novo transcriptome assembly.
FastA: Genome-guided transcriptome assembly.

PASA:

GFF3: Genome annotations produced by PASA pipeline in GFF3.
BED: Genome annotations produced by PASA pipeline in BED format.

2 SETUP

2.1 Libraries and markdown settings

library(knitr)
library(reticulate)
knitr::opts_chunk$set(
  echo = TRUE,         # Display code chunks
  eval = FALSE,        # Evaluate code chunks
  warning = FALSE,     # Hide warnings
  message = FALSE,     # Hide messages
  comment = ""         # Prevents appending '##' to beginning of lines in code output
)

2.2 Set variables

# DIRECTORIES
top_output_dir <- file.path("..", "output")

output_dir <- file.path(top_output_dir, "00.30-E-Peve-transcriptome-assembly-Trinity")
de_novo_output_dir <- file.path(output_dir, "de_novo_assembly")
genome_guided_output_dir <- file.path(output_dir, "genome_guided_assembly")
pasa_container_dir <- file.path("/home", "shared", "containers")
pasa_bed <- "peve-PASA.bed"
pasa_gff <- "peve-PASA.gff3"
PASA_HOME <- "/usr/local/src/PASApipeline"
pasa_output_dir <- file.path(output_dir, "PASA")
stringtie_gtf_dir <- file.path(top_output_dir, "02.20-E-Peve-RNAseq-alignment-HiSat2")
trimmed_reads_dir <- file.path(top_output_dir, "01.00-E-Peve-RNAseq-trimming-fastp-FastQC-MultiQC")


# FILES
bam_alignment <- file.path(top_output_dir, "02.20-E-Peve-RNAseq-alignment-HiSat2", "sorted-bams-merged.bam")

## Path for genome will be relative to PASA output dir
genome_fasta <- file.path("..", "..", "..", "data", "Porites_evermanni_v1.fa")
genome_gff <- file.path("..", "data", "Porites_evermanni_validated.gff3")
denovo_assembly_name <- "peve-denovo-Trinity"
genome_guided_assembly_name <- "peve-GG-Trinity"
pasa_container <- "pasapipeline.v2.5.3.simg"
stringtie_gtf <- file.path(stringtie_gtf_dir, "Porites_evermanni_v1.stringtie.gtf")

#SETTINGS
## THREADS
threads <- "44"

## MAX RAM
max_ram <- "100G"

# PROGRAMS
samtools <- file.path("/home", "shared", "samtools-1.12", "samtools")


# FORMATTING
line <- "-----------------------------------------------"

# Export these as environment variables for bash chunks.
Sys.setenv(
  bam_alignment = bam_alignment,
  denovo_assembly_name = denovo_assembly_name,
  de_novo_output_dir = de_novo_output_dir,
  genome_fasta = genome_fasta,
  genome_gff = genome_gff,
  genome_guided_assembly_name = genome_guided_assembly_name,
  genome_guided_output_dir = genome_guided_output_dir,
  line = line,
  max_ram = max_ram,
  output_dir = output_dir,
  top_output_dir = top_output_dir,
  pasa_bed = pasa_bed,
  pasa_container = pasa_container,
  pasa_container_dir = pasa_container_dir,
  pasa_gff = pasa_gff,
  PASA_HOME = PASA_HOME,
  pasa_output_dir = pasa_output_dir,
  samtools = samtools,
  stringtie_gtf_dir = stringtie_gtf_dir,
  stringtie_gtf = stringtie_gtf,
  threads = threads,
  trimmed_reads_dir = trimmed_reads_dir
)

3 DE NOVO ASSEMBLY

3.1 Run Trinity

Trinity was run using the Trinity Singularity (Apptainer) container, trinityrnaseq.v2.15.2.simg from the urol-e5/timeseries_molecular/E-Peve/code/ directory.

This was done in a terminal, outside of this notebook.

3.1.1 Set Bash variables

# Directories
top_output_dir="../output"

output_dir="${top_output_dir}/00.30-E-Peve-transcriptome-assembly-Trinity"
de_novo_output_dir="${output_dir}/de_novo_assembly"
genome_guided_output_dir="${output_dir}/genome_guided_assembly"
trimmed_reads_dir="${top_output_dir}/01.00-E-Peve-RNAseq-trimming-fastp-FastQC-MultiQC"

# PASA INPUT FILES
####### NEED TO BE RELATIVE TO PASA SUBDIRECTORY #######
genome_fasta="../../../data/Porites_evermanni_v1.fa"
genome_gff="../../../data/Porites_evermanni_validated.gff3"
pasa_container="pasapipeline.v2.5.3.simg"
PASA_HOME="/usr/local/src/PASApipeline"
stringtie_gtf="../../02.20-E-Peve-RNAseq-alignment-HiSat2/Porites_evermanni_v1.stringtie.gtf"

## THREADS
threads="44"

## MAX RAM
max_ram="100G"

# Make output directoy, if it doesn't exist
mkdir --parents ${de_novo_output_dir}

## Inititalize arrays
R1_array=()
R2_array=()

# Variables for R1/R2 lists
R1_list=""
R2_list=""

# Create array of fastq R1 files
R1_array=(${trimmed_reads_dir}/*R1_001.fastp-trim.fq.gz)

# Create array of fastq R2 files
R2_array=(${trimmed_reads_dir}/*R2_001.fastp-trim.fq.gz)

# Create list of fastq files used in analysis
## Uses parameter substitution to strip leading path from filename
if [ ! -f "${de_novo_output_dir}/fastq.list.txt" ]; then
  for fastq in ${trimmed_reads_dir}/*.fq.gz
  do
    echo "${fastq##*/}" >> ${de_novo_output_dir}/fastq.list.txt
  done
fi

# Create comma-separated lists of FastQ reads
R1_list=$(echo "${R1_array[@]}" | tr " " ",")
R2_list=$(echo "${R2_array[@]}" | tr " " ",")

3.1.2 Run Trinity Singularity image.

Used “stranded” setting (–SS_lib_type).

singularity exec \
-B /home \
-e trinityrnaseq.v2.15.2.simg \
Trinity \
--seqType fq \
--max_memory ${max_ram} \
--CPU ${threads} \
--SS_lib_type RF \
--left "${R1_list}" \
--right "${R2_list}" \
--output ${de_novo_output_dir}/trinity_out_dir \
--full_cleanup \
> ${de_novo_output_dir}/trinity.log \
2>&1

3.2 Rename output files

3.2.1 Rename FastA

# Rename generic assembly FastA
mv ${de_novo_output_dir}/trinity_out_dir.Trinity.fasta \
${de_novo_output_dir}/${denovo_assembly_name}.fasta

3.2.2 Rename gene map and log

mv ${de_novo_output_dir}/trinity_out_dir.Trinity.fasta.gene_trans_map \
${de_novo_output_dir}/${denovo_assembly_name}.gene_trans_map

mv ${de_novo_output_dir}/trinity.log \
${de_novo_output_dir}/${denovo_assembly_name}.log

3.2.3 Assembly stats

3.2.3.1 Run Trinity Singularity image.

singularity exec -B /home \
-e trinityrnaseq.v2.15.2.simg \
/usr/local/bin/util/TrinityStats.pl \
../output/00.30-E-Peve-transcriptome-assembly-Trinity/de_novo_assembly/peve-denovo-Trinity.fasta \
> ../output/00.30-E-Peve-transcriptome-assembly-Trinity/de_novo_assembly/peve-denovo-Trinity.stats

3.3 Create FastA index

${samtools} faidx \
${de_novo_output_dir}/${denovo_assembly_name}.fasta

3.4 Checksums

cd ${de_novo_output_dir}

md5sum ${denovo_assembly_name}.fasta | tee ${denovo_assembly_name}.fasta.md5

72835974384b1693cad623628ae9315c  peve-denovo-Trinity.fasta

4 GENOME-GUIDED ASSEMBLY

Trinity was run using the Trinity Singularity (Apptainer) container, trinityrnaseq.v2.15.2.simg from the urol-e5/timeseries_molecular/E-Peve/code/ directory.

This was done in a terminal, outside of this notebook.

singularity exec \
-B /home -e trinityrnaseq.v2.15.2.simg \
Trinity \
--genome_guided_bam ../output/02.20-E-Peve-RNAseq-alignment-HiSat2/sorted-bams-merged.bam \
--genome_guided_max_intron 10000 \
--max_memory ${max_ram} \
--CPU ${threads} \
--SS_lib_type RF \
--output ${genome_guided_output_dir}/trinity_out_dir \
--full_cleanup \
> ${genome_guided_output_dir}/trinity.log 2>&1

4.1 Rename output files

4.1.1 Rename FastA

# Rename generic assembly FastA
mv ${genome_guided_output_dir}/trinity_out_dir.Trinity-GG.fasta \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta

4.1.2 Rename gene map and log

mv ${genome_guided_output_dir}/trinity_out_dir.Trinity-GG.fasta.gene_trans_map \
${genome_guided_output_dir}/${genome_guided_assembly_name}.gene_trans_map

mv ${genome_guided_output_dir}/trinity.log \
${genome_guided_output_dir}/${genome_guided_assembly_name}.log

4.2 Create FastA index

${samtools} faidx \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta

4.3 Checksums

cd ${genome_guided_output_dir}

md5sum ${genome_guided_assembly_name}.fasta | tee ${genome_guided_assembly_name}.fasta.md5

d60d91acf9621a424fe61c1dbfc2d875  peve-GG-Trinity.fasta

4.3.1 Assembly stats

4.3.1.1 Run Trinity Singularity image.

singularity exec -B /home \
-e trinityrnaseq.v2.15.2.simg \
/usr/local/bin/util/TrinityStats.pl \
../output/00.30-E-Peve-transcriptome-assembly-Trinity/genome_guided_assembly/peve-GG-Trinity.fasta \
> ../output/00.30-E-Peve-transcriptome-assembly-Trinity/genome_guided_assembly/peve-GG-Trinity.stats

5 PASA PIPELINE

5.1 Concatenate Trinity assemblies

cat ${de_novo_output_dir}/${denovo_assembly_name}.fasta \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta \
> ${pasa_output_dir}/transcripts.fasta

5.1.1 Confirm counts

# Count transcripts in each file
denovo_count=$(grep -c "^>" ${de_novo_output_dir}/${denovo_assembly_name}.fasta)
genome_guided_count=$(grep -c "^>" ${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta)
pasa_count=$(grep -c "^>" ${pasa_output_dir}/transcripts.fasta)

# Calculate sum of first two counts
sum=$((denovo_count + genome_guided_count))

# Compare sum to PASA count
echo "De novo count: $denovo_count"
echo "Genome-guided count: $genome_guided_count"
echo "Sum: $sum"
echo "PASA count: $pasa_count"

if [ $sum -eq $pasa_count ]; then
    echo "✓ Counts match: $sum = $pasa_count"
else
    echo "✗ Counts do not match: $sum ≠ $pasa_count (difference: $((pasa_count - sum)))"
fi

De novo count: 1065667
Genome-guided count: 533664
Sum: 1599331
PASA count: 1599331
✓ Counts match: 1599331 = 1599331

5.2 Extract transcript accessions

singularity exec \
-B /home \
-e ${pasa_container_dir}/${pasa_container} \
$PASA_HOME/misc_utilities/accession_extractor.pl \
< ${de_novo_output_dir}/${denovo_assembly_name}.fasta \
> ${pasa_output_dir}/tdn.accs

head ${pasa_output_dir}/tdn.accs

5.3 Clean transcripts

cd ${pasa_output_dir}

singularity exec \
-B /home \
-e \
--env USER="$USER" \
${pasa_container} \
$PASA_HOME/bin/seqclean \
transcripts.fasta \
-c 16

5.4 PASA Assembly

5.4.1 Fix schema key length

cd ${pasa_output_dir}

#### Fix schema key length issue ####
singularity exec ${pasa_container} \
cat /usr/local/src/PASApipeline/schema/cdna_alignment_mysqlschema \
> cdna_alignment_mysqlschema

# Fix all variations of gene_id and model_id indexes
sed -i 's/KEY gene_id_idx (gene_id)/KEY gene_id_idx (gene_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/KEY mod_idx (model_id)/KEY mod_idx (model_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/(gene_id)/(gene_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/(model_id)/(model_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/KEY gene_idx (annotation_version,gene_id)/KEY gene_idx (annotation_version,gene_id(255))/g' cdna_alignment_mysqlschema

5.4.2 Run PASA Assembly Pipeline

This was executed outside of RStudio due to the verbose output, which will cause RStudio to crash.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
--config alignAssembly.config \
--create \
--run \
--genome ${genome_fasta} \
--transcripts transcripts.fasta.clean \
--trans_gtf ${stringtie_gtf} \
--ALT_SPLICE \
-T \
-u transcripts.fasta \
--ALIGNERS blat,gmap,minimap2 \
--TDN tdn.accs \
--transcribed_is_aligned_orient \
--annot_compare \
-L \
--annots ${genome_gff} \
--TRANSDECODER \
--CPU ${threads}

5.4.3 Alternative Splicing

This doesn’t seem to have run during the assembly phase, so ran separately.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
-c alignAssembly.config \
--ALT_SPLICE \
-g ${genome_fasta} \
-t all.transcripts.fasta.clean \
--CPU ${threads}

5.4.4 Update annotations

Now includes alternative splicing info.

Uses output GFF3 from initial annotations as annotation input.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
-c annotCompare.config \
--annot_compare \
-L \
--annots peve_pasa.gene_structures_post_PASA_updates.2550221.gff3 \
-g ${genome_fasta} \
-t all.transcripts.fasta.clean \
--CPU ${threads}

6 PASA OUTPUTS

6.1 Generate checksums

cd "${pasa_output_dir}"

md5sum peve_pasa.gene_structures_post_PASA_updates.3761502.gff3 | tee peve_pasa.gene_structures_post_PASA_updates.3761502.gff3.md5

md5sum peve_pasa.gene_structures_post_PASA_updates.3761502.bed | tee peve_pasa.gene_structures_post_PASA_updates.3761502.bed.md5

08d65a59913b56823d9a108db722f820  peve_pasa.gene_structures_post_PASA_updates.3761502.gff3
b4f76f4651063be32724c86d067ad34a  peve_pasa.gene_structures_post_PASA_updates.3761502.bed

6.2 Rename outputs

cd "${pasa_output_dir}"
cp peve_pasa.gene_structures_post_PASA_updates.3761502.gff3 "${pasa_gff}"

cp peve_pasa.gene_structures_post_PASA_updates.3761502.bed "${pasa_bed}"

md5sum "${pasa_gff}" | tee "${pasa_gff}".md5
md5sum "${pasa_bed}" | tee "${pasa_bed}".md5

08d65a59913b56823d9a108db722f820  peve-PASA.gff3
b4f76f4651063be32724c86d067ad34a  peve-PASA.bed

6.3 GFF3 Preview

head -n 50 "${pasa_output_dir}"/"${pasa_gff}"

# PASA_UPDATE: mrna-26667, single gene model update, valid-1, status:[pasa:asmbl_337815,status:12], valid-1
Porites_evermani_scaffold_4224  .   gene    28057   29953   .   -   .   ID=gene-Peve_00030015;Name=mrna-26667
Porites_evermani_scaffold_4224  .   mRNA    28057   29953   .   -   .   ID=mrna-26667;Parent=gene-Peve_00030015;Name=mrna-26667
Porites_evermani_scaffold_4224  .   five_prime_UTR  29913   29953   .   -   .   ID=mrna-26667.utr5p1;Parent=mrna-26667
Porites_evermani_scaffold_4224  .   exon    29544   29953   .   -   .   ID=mrna-26667.exon1;Parent=mrna-26667
Porites_evermani_scaffold_4224  .   CDS 29544   29912   .   -   0   ID=mrna-26667.cds.1;Parent=mrna-26667
Porites_evermani_scaffold_4224  .   exon    28057   29172   .   -   .   ID=mrna-26667.exon2;Parent=mrna-26667
Porites_evermani_scaffold_4224  .   CDS 29008   29172   .   -   0   ID=mrna-26667.cds.2;Parent=mrna-26667
Porites_evermani_scaffold_4224  .   three_prime_UTR 28057   29007   .   -   .   ID=mrna-26667.utr3p1;Parent=mrna-26667


#PROT mrna-26667 gene-Peve_00030015 MPGPRGDRGREGPPGKSGPRGINGMKGQQGLLGMKGEPGIAGMKGERGAVGIKGQRGIVGMKGVPGAVGIKGERGIVGNPGRKGNKGEKGESAKASQASVVPQTNWKQCVWNSDSGTDNGKIKDCAFNKLQSNSALKVSFQGNMRVIGHGACNRWYFKFNGNECNGPMTIEAIVNNH*

# PASA_UPDATE: novel_model_1794_698a2f3c, single gene model update, valid-1, status:[pasa:asmbl_411975,status:3], valid-1
Porites_evermani_scaffold_594   .   gene    170578  170986  .   +   .   ID=novel_gene_1397_698a2f3c;Name=%2A%2A%20NO%20NAME%20ASSIGNED%20%2A%2A
Porites_evermani_scaffold_594   .   mRNA    170578  170986  .   +   .   ID=novel_model_1794_698a2f3c;Parent=novel_gene_1397_698a2f3c;Name=%2A%2A%20NO%20NAME%20ASSIGNED%20%2A%2A
Porites_evermani_scaffold_594   .   five_prime_UTR  170578  170640  .   +   .   ID=novel_model_1794_698a2f3c.utr5p1;Parent=novel_model_1794_698a2f3c
Porites_evermani_scaffold_594   .   exon    170578  170986  .   +   .   ID=novel_model_1794_698a2f3c.exon1;Parent=novel_model_1794_698a2f3c
Porites_evermani_scaffold_594   .   CDS 170641  170952  .   +   0   ID=novel_model_1794_698a2f3c.cds.1;Parent=novel_model_1794_698a2f3c
Porites_evermani_scaffold_594   .   three_prime_UTR 170953  170986  .   +   .   ID=novel_model_1794_698a2f3c.utr3p1;Parent=novel_model_1794_698a2f3c


#PROT novel_model_1794_698a2f3c novel_gene_1397_698a2f3c    MSKTPLQDEYLQWRPQILGITTQLNAFFFNHLMKSCHPFKSHATEYLCSTCNSGVCSDPPLPFEAWVKCTMYIDPTLSKIDTFGTSTKCPSKSDVCLLESQKM*

# ORIGINAL: mrna-32793 original gene structure, not modified by PASA
Porites_evermani_scaffold_594   Gmove   gene    82339   84386   .   +   .   ID=gene-Peve_00036817;Name=mrna-32793
Porites_evermani_scaffold_594   Gmove   mRNA    82339   84386   .   +   .   ID=mrna-32793;Parent=gene-Peve_00036817;Name=mrna-32793
Porites_evermani_scaffold_594   Gmove   exon    82339   82462   .   +   .   ID=mrna-32793.exon1;Parent=mrna-32793
Porites_evermani_scaffold_594   Gmove   CDS 82339   82462   .   +   0   ID=mrna-32793.cds.1;Parent=mrna-32793
Porites_evermani_scaffold_594   Gmove   exon    83270   83388   .   +   .   ID=mrna-32793.exon2;Parent=mrna-32793
Porites_evermani_scaffold_594   Gmove   CDS 83270   83388   .   +   2   ID=mrna-32793.cds.2;Parent=mrna-32793
Porites_evermani_scaffold_594   Gmove   exon    84354   84386   .   +   .   ID=mrna-32793.exon3;Parent=mrna-32793
Porites_evermani_scaffold_594   Gmove   CDS 84354   84386   .   +   0   ID=mrna-32793.cds.3;Parent=mrna-32793


#PROT mrna-32793 gene-Peve_00036817 LKAIVTTEGVYFVRARGTPSPESIVLLVRFSDLYVCQPLSTDGKSYIELVMKADNSSVPTPSKNPSKRPRVRCDQQSVAQKVAQKINYAKNL

# PASA_UPDATE: mrna-32790, single gene model update, valid-1, status:[pasa:asmbl_411993,status:12], valid-1
Porites_evermani_scaffold_594   .   gene    203517  221865  .   +   .   ID=gene-Peve_00036814;Name=mrna-32790
Porites_evermani_scaffold_594   .   mRNA    203517  221865  .   +   .   ID=mrna-32790;Parent=gene-Peve_00036814;Name=mrna-32790
Porites_evermani_scaffold_594   .   exon    203517  203643  .   +   .   ID=mrna-32790.exon1;Parent=mrna-32790
Porites_evermani_scaffold_594   .   CDS 203517  203643  .   +   0   ID=mrna-32790.cds.1;Parent=mrna-32790
Porites_evermani_scaffold_594   .   exon    211961  212176  .   +   .   ID=mrna-32790.exon2;Parent=mrna-32790
Porites_evermani_scaffold_594   .   CDS 211961  212176  .   +   2   ID=mrna-32790.cds.2;Parent=mrna-32790
Porites_evermani_scaffold_594   .   exon    217811  217948  .   +   .   ID=mrna-32790.exon3;Parent=mrna-32790
Porites_evermani_scaffold_594   .   CDS 217811  217948  .   +   2   ID=mrna-32790.cds.3;Parent=mrna-32790
Porites_evermani_scaffold_594   .   exon    219213  219368  .   +   .   ID=mrna-32790.exon4;Parent=mrna-32790
Porites_evermani_scaffold_594   .   CDS 219213  219368  .   +   2   ID=mrna-32790.cds.4;Parent=mrna-32790
Porites_evermani_scaffold_594   .   exon    220000  220155  .   +   .   ID=mrna-32790.exon5;Parent=mrna-32790
Porites_evermani_scaffold_594   .   CDS 220000  220155  .   +   2   ID=mrna-32790.cds.5;Parent=mrna-32790

6.4 GFF Comparisons

printf '%s\n\n' "Original GFF feature counts:"
awk '!/^#/ && !/^[[:space:]]*$/ && NF > 0 && $3 != "" {print $3}' ${genome_gff} \
| sort | uniq -c | sort -rn | awk '{print $2, $1}'

echo ""
echo "${line}"
echo ""

printf "%s\n\n" "Updated GFF feature counts:"
awk -F "\t" '!/^#/ && !/^[[:space:]]*$/ && NF > 0 && $3 != "" {print $3}' "${pasa_output_dir}"/"${pasa_gff}" \
| sort | uniq -c | sort -rn | awk '{print $2, $1}'

Original GFF feature counts:

exon 246418
CDS 231320
mRNA 40389
gene 40389
UTR 15098

-----------------------------------------------

Updated GFF feature counts:

exon 463084
CDS 442258
mRNA 59418
five_prime_UTR 43027
gene 40808
three_prime_UTR 38919

7 EXTRACT PROTEINS TO FASTA

cd "${pasa_output_dir}"

awk '/^#PROT / {print ">" $2 "." $3 "\n" $4}' "${pasa_gff}" > peve-proteins-PASA.fasta

printf "%s\n\n" "Original protein counts:"
grep --count "^#PROT" "${pasa_gff}"

echo ""
echo "${line}"
echo ""

printf "%s\n\n" "Extracted protein counts:"
grep --count "^>" peve-proteins-PASA.fasta

# Create FastA Index
${samtools} faidx peve-proteins-PASA.fasta

Original protein counts:

59418

-----------------------------------------------

Extracted protein counts:

59418

7.1 Checksums

cd "${pasa_output_dir}"
md5sum peve-proteins-PASA.fasta | tee peve-proteins-PASA.fasta.md5

9198a4d1e67759b0a78dd4d9f4229695  peve-proteins-PASA.fasta

De Novo Transcriptome Assembly and Annotation - E5 P.tuahiensis RNA-seq Using Trinity and PASA

Sam White — Fri, 27 Feb 2026 08:00:00 GMT

INTRO

This notebook performs a comprehensive transcriptome assembly and annotation for P.tuahiensis using the PASA (Program to Assemble Spliced Alignments), as part of the E5 timeseries_molecular project. This produced an updated genome annotation that incorporates the RNA-seq data, including alternative splicing information. However, it’s important to note that the resulting GFF/BED files will not contain any annotations from the original genome GFF which did not have any support from RNA-seq alignments!

Due to large file sizes, the majority of important output files are not availabe on GitHub, but can be accessed on Gannet here:

https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/F-Ptua/output/00.30-F-Ptua-transcriptome-assembly-Trinity/

Primary products:

Trinity de-novo assembly FastA: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/F-Ptua/output/00.30-F-Ptua-transcriptome-assembly-Trinity/de_novo_assembly/ptua-denovo-Trinity.fasta
Trinity genome-guided assembly FastA: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/F-Ptua/output/00.30-F-Ptua-transcriptome-assembly-Trinity/genome_guided_assembly/ptua-GG-Trinity.fasta
PASA final GFF: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/F-Ptua/output/00.30-F-Ptua-transcriptome-assembly-Trinity/PASA/ptua-PASA.gff3
PASA final BED: https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/F-Ptua/output/00.30-F-Ptua-transcriptome-assembly-Trinity/PASA/ptua-PASA.bed

The markdown below was produced from the original Rmd script:

00.30-F-Ptua-transcriptome-assembly-Trinity.Rmd (GitHub)

Not described below is the access to the PASA web portal, which allows the user to browse the PASA results in a user-friendly interface. This can be accessed here:

http://gannet.fish.washington.edu:9000/

Enter ptua_pasa as the database name. If this is your first time using the database, it will take many minutes for the data to load (the MySQL files behind the scenes are ~10GB in size, so it takes a while to process). After initial acces, the data is cached and access is much faster.

Screenshot of ptua_pasa web portal on Gannet

1 BACKGROUND

This notebook performs a comprehensive transcriptome assembly and annotaiton for P.tuahiensis using the PASA (Program to Assemble Spliced Alignments) pipeline, as well as alternative isoform identification. This pipeline relies on both de novo and genome-guided transcriptome assemblies with Trinity.

1.1 Expected outputs

TRINITY:

FastA: De novo transcriptome assembly.
FastA: Genome-guided transcriptome assembly.

PASA:

GFF3: Genome annotations produced by PASA pipeline in GFF3.
BED: Genome annotations produced by PASA pipeline in BED format.

2 SETUP

2.1 Libraries and markdown settings

library(knitr)
library(reticulate)
knitr::opts_chunk$set(
  echo = TRUE,         # Display code chunks
  eval = FALSE,        # Evaluate code chunks
  warning = FALSE,     # Hide warnings
  message = FALSE,     # Hide messages
  comment = ""         # Prevents appending '##' to beginning of lines in code output
)

2.2 Set variables

# DIRECTORIES
top_output_dir <- file.path("..", "output")

output_dir <- file.path(top_output_dir, "00.30-F-Ptua-transcriptome-assembly-Trinity")
de_novo_output_dir <- file.path(output_dir, "de_novo_assembly")
genome_guided_output_dir <- file.path(output_dir, "genome_guided_assembly")
pasa_container_dir <- file.path("/home", "shared", "containers")
PASA_HOME <- "/usr/local/src/PASApipeline"
pasa_output_dir <- file.path(output_dir, "PASA")
stringtie_gtf_dir <- file.path(top_output_dir, "02.20-F-Ptua-RNAseq-alignment-HiSat2")
trimmed_reads_dir <- file.path(top_output_dir, "01.00-F-Ptua-RNAseq-trimming-fastp-FastQC-MultiQC")


# FILES
bam_alignment <- file.path(top_output_dir, "02.20-F-Ptua-RNAseq-alignment-HiSat2", "sorted-bams-merged.bam")

## Path for genome will be relative to PASA output dir
genome_fasta <- file.path("..", "..", "..", "data", "Pocillopora_meandrina_HIv1.assembly.fasta")
genome_gff <- file.path("..", "data", "Pocillopora_meandrina_HIv1.genes.gff3")
denovo_assembly_name <- "ptua-denovo-Trinity"
genome_guided_assembly_name <- "ptua-GG-Trinity"
pasa_bed <- "ptua-PASA.bed"
pasa_container <- "pasapipeline.v2.5.3.simg"
pasa_gff <- "ptua-PASA.gff3"
stringtie_gtf <- file.path(stringtie_gtf_dir, "Pocillopora_meandrina_HIv1.assembly.stringtie.gtf")

#SETTINGS
## THREADS
threads <- "44"

## MAX RAM
max_ram <- "100G"

# PROGRAMS
samtools <- file.path("/home", "shared", "samtools-1.12", "samtools")


# FORMATTING
line <- "-----------------------------------------------"

# Export these as environment variables for bash chunks.
Sys.setenv(
  bam_alignment = bam_alignment,
  denovo_assembly_name = denovo_assembly_name,
  de_novo_output_dir = de_novo_output_dir,
  genome_fasta = genome_fasta,
  genome_gff = genome_gff,
  genome_guided_assembly_name = genome_guided_assembly_name,
  genome_guided_output_dir = genome_guided_output_dir,
  line = line,
  max_ram = max_ram,
  output_dir = output_dir,
  top_output_dir = top_output_dir,
  pasa_bed = pasa_bed,
  pasa_container = pasa_container,
  pasa_container_dir = pasa_container_dir,
  pasa_gff = pasa_gff,
  PASA_HOME = PASA_HOME,
  pasa_output_dir = pasa_output_dir,
  samtools = samtools,
  stringtie_gtf_dir = stringtie_gtf_dir,
  stringtie_gtf = stringtie_gtf,
  threads = threads,
  trimmed_reads_dir = trimmed_reads_dir
)

3 DE NOVO ASSEMBLY

3.1 Run Trinity

Trinity was run using the Trinity Singularity (Apptainer) container, trinityrnaseq.v2.15.2.simg from the urol-e5/timeseries_molecular/F-Ptua/code/ directory.

This was done in a terminal, outside of this notebook.

3.1.1 Set Bash variables

# Directories
top_output_dir="../output"

output_dir="${top_output_dir}/00.30-F-Ptua-transcriptome-assembly-Trinity"
de_novo_output_dir="${output_dir}/de_novo_assembly"
genome_guided_output_dir="${output_dir}/genome_guided_assembly"
pasa_output_dir="${output_dir}/PASA"
trimmed_reads_dir="${top_output_dir}/01.00-F-Ptua-RNAseq-trimming-fastp-FastQC-MultiQC"

# FILES
bam_alignment="${top_output_dir}/02.20-F-Ptua-RNAseq-alignment-HiSat2/sorted-bams-merged.bam"
denovo_assembly_name="ptua-denovo-Trinity"
genome_guided_assembly_name="ptua-GG-Trinity"

# PASA INPUT FILES
####### NEED TO BE RELATIVE TO PASA SUBDIRECTORY #######
genome_fasta="../../../data/Pocillopora_meandrina_HIv1.assembly.fasta"
genome_gff="../../../data/Pocillopora_meandrina_HIv1.genes-validated.gff3"
pasa_container="pasapipeline.v2.5.3.simg"
PASA_HOME="/usr/local/src/PASApipeline"
stringtie_gtf="../../../output/02.20-F-Ptua-RNAseq-alignment-HiSat2/Pocillopora_meandrina_HIv1.assembly.stringtie.gtf 
"

## THREADS
threads="44"

## MAX RAM
max_ram="100G"

# Make output directoy, if it doesn't exist
mkdir --parents ${de_novo_output_dir}
mkdir --parents ${pasa_output_dir}

## Inititalize arrays
R1_array=()
R2_array=()

# Variables for R1/R2 lists
R1_list=""
R2_list=""

# Create array of fastq R1 files
R1_array=(${trimmed_reads_dir}/*R1_001.fastp-trim.fq.gz)

# Create array of fastq R2 files
R2_array=(${trimmed_reads_dir}/*R2_001.fastp-trim.fq.gz)

# Create list of fastq files used in analysis
## Uses parameter substitution to strip leading path from filename
if [ ! -f "${de_novo_output_dir}/fastq.list.txt" ]; then
  for fastq in ${trimmed_reads_dir}/*.fq.gz
  do
    echo "${fastq##*/}" >> ${de_novo_output_dir}/fastq.list.txt
  done
fi

# Create comma-separated lists of FastQ reads
R1_list=$(echo "${R1_array[@]}" | tr " " ",")
R2_list=$(echo "${R2_array[@]}" | tr " " ",")

3.1.2 Run Trinity Singularity image.

Used “stranded” setting (–SS_lib_type).

singularity exec \
-B /home \
-e trinityrnaseq.v2.15.2.simg \
Trinity \
--seqType fq \
--max_memory ${max_ram} \
--CPU ${threads} \
--SS_lib_type RF \
--left "${R1_list}" \
--right "${R2_list}" \
--output ${de_novo_output_dir}/trinity_out_dir \
--full_cleanup \
> ${de_novo_output_dir}/trinity.log \
2>&1

3.2 Rename output files

3.2.1 Rename FastA

# Rename generic assembly FastA
mv ${de_novo_output_dir}/trinity_out_dir.Trinity.fasta \
${de_novo_output_dir}/${denovo_assembly_name}.fasta

3.2.2 Rename gene map and log

mv ${de_novo_output_dir}/trinity_out_dir.Trinity.fasta.gene_trans_map \
${de_novo_output_dir}/${denovo_assembly_name}.gene_trans_map

mv ${de_novo_output_dir}/trinity.log \
${de_novo_output_dir}/${denovo_assembly_name}.log

3.2.3 Assembly stats

3.2.3.1 Run Trinity Singularity image.

singularity exec -B /home \
-e trinityrnaseq.v2.15.2.simg \
/usr/local/bin/util/TrinityStats.pl \
../output/00.30-F-Ptua-transcriptome-assembly-Trinity/de_novo_assembly/ptua-denovo-Trinity.fasta \
> ../output/00.30-F-Ptua-transcriptome-assembly-Trinity/de_novo_assembly/ptua-denovo-Trinity.stats

3.3 Create FastA index

${samtools} faidx \
${de_novo_output_dir}/${denovo_assembly_name}.fasta

3.4 Checksums

cd ${de_novo_output_dir}

md5sum ${denovo_assembly_name}.fasta | tee ${denovo_assembly_name}.fasta.md5

0c3073b02a10eb38893e296ce9d9dfb3  ptua-denovo-Trinity.fasta

4 GENOME-GUIDED ASSEMBLY

Trinity was run using the Trinity Singularity (Apptainer) container, trinityrnaseq.v2.15.2.simg from the urol-e5/timeseries_molecular/F-Ptua/code/ directory.

This was done in a terminal, outside of this notebook.

singularity exec \
-B /home -e trinityrnaseq.v2.15.2.simg \
Trinity \
--genome_guided_bam ${bam_alignment} \
--genome_guided_max_intron 10000 \
--max_memory ${max_ram} \
--CPU ${threads} \
--SS_lib_type RF \
--output ${genome_guided_output_dir}/trinity_out_dir \
--full_cleanup \
> ${genome_guided_output_dir}/trinity.log 2>&1

4.1 Rename output files

4.1.1 Rename FastA

# Rename generic assembly FastA
mv ${genome_guided_output_dir}/trinity_out_dir.Trinity-GG.fasta \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta

4.1.2 Rename gene map and log

mv ${genome_guided_output_dir}/trinity_out_dir.Trinity-GG.fasta.gene_trans_map \
${genome_guided_output_dir}/${genome_guided_assembly_name}.gene_trans_map

mv ${genome_guided_output_dir}/trinity.log \
${genome_guided_output_dir}/${genome_guided_assembly_name}.log

4.2 Create FastA index

${samtools} faidx \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta

4.3 Checksums

cd ${de_novo_output_dir}

md5sum ${denovo_assembly_name}.fasta | tee ${denovo_assembly_name}.fasta.md5

0c3073b02a10eb38893e296ce9d9dfb3  ptua-denovo-Trinity.fasta

4.3.1 Assembly stats

4.3.1.1 Run Trinity Singularity image.

singularity exec -B /home \
-e trinityrnaseq.v2.15.2.simg \
/usr/local/bin/util/TrinityStats.pl \
../output/00.30-F-Ptua-transcriptome-assembly-Trinity/genome_guided_assembly/ptua-GG-Trinity.fasta \
> ../output/00.30-F-Ptua-transcriptome-assembly-Trinity/genome_guided_assembly/ptua-GG-Trinity.stats

5 PASA PIPELINE

5.1 Concatenate Trinity assemblies

cat ${de_novo_output_dir}/${denovo_assembly_name}.fasta \
${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta \
> ${pasa_output_dir}/transcripts.fasta

5.1.1 Confirm counts

# Count transcripts in each file
denovo_count=$(grep -c "^>" ${de_novo_output_dir}/${denovo_assembly_name}.fasta)
genome_guided_count=$(grep -c "^>" ${genome_guided_output_dir}/${genome_guided_assembly_name}.fasta)
pasa_count=$(grep -c "^>" ${pasa_output_dir}/transcripts.fasta)

# Calculate sum of first two counts
sum=$((denovo_count + genome_guided_count))

# Compare sum to PASA count
echo "De novo count: $denovo_count"
echo "Genome-guided count: $genome_guided_count"
echo "Sum: $sum"
echo "PASA count: $pasa_count"

if [ $sum -eq $pasa_count ]; then
    echo "✓ Counts match: $sum = $pasa_count"
else
    echo "✗ Counts do not match: $sum ≠ $pasa_count (difference: $((pasa_count - sum)))"
fi

De novo count: 1232241
Genome-guided count: 499541
Sum: 1731782
PASA count: 1731782
✓ Counts match: 1731782 = 1731782

5.2 Extract transcript accessions

singularity exec \
-B /home \
-e ${pasa_container_dir}/${pasa_container} \
$PASA_HOME/misc_utilities/accession_extractor.pl \
< ${de_novo_output_dir}/${denovo_assembly_name}.fasta \
> ${pasa_output_dir}/tdn.accs

head ${pasa_output_dir}/tdn.accs

5.3 Clean transcripts

cd ${pasa_output_dir}

singularity exec \
-B /home \
-e \
--env USER="$USER" \
${pasa_container} \
$PASA_HOME/bin/seqclean \
transcripts.fasta \
-c 16

5.4 PASA Assembly

5.4.1 Fix schema key length

cd ${pasa_output_dir}

#### Fix schema key length issue ####
singularity exec ${pasa_container} \
cat /usr/local/src/PASApipeline/schema/cdna_alignment_mysqlschema \
> cdna_alignment_mysqlschema

# Fix all variations of gene_id and model_id indexes
sed -i 's/KEY gene_id_idx (gene_id)/KEY gene_id_idx (gene_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/KEY mod_idx (model_id)/KEY mod_idx (model_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/(gene_id)/(gene_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/(model_id)/(model_id(255))/g' cdna_alignment_mysqlschema
sed -i 's/KEY gene_idx (annotation_version,gene_id)/KEY gene_idx (annotation_version,gene_id(255))/g' cdna_alignment_mysqlschema

5.4.2 Run PASA Assembly Pipeline

This was executed outside of RStudio due to the verbose output, which will cause RStudio to crash.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
--config alignAssembly.config \
--create \
--run \
--genome ${genome_fasta} \
--transcripts transcripts.fasta.clean \
--trans_gtf ${stringtie_gtf} \
--ALT_SPLICE \
-T \
-u transcripts.fasta \
--ALIGNERS blat,gmap,minimap2 \
--TDN tdn.accs \
--transcribed_is_aligned_orient \
--annot_compare \
-L \
--annots ${genome_gff} \
--TRANSDECODER \
--CPU ${threads}

5.4.3 Alternative Splicing

This doesn’t seem to have run during the assembly phase, so ran separately.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
-c alignAssembly.config \
--ALT_SPLICE \
-g ${genome_fasta} \
-t all.transcripts.fasta.clean \
--CPU ${threads}

5.4.4 Update annotations

Now includes alternative splicing info.

Uses output GFF3 from initial annotations as annotation input.

singularity exec \
-B /home \
-B /var/run/mysqld/mysqld.sock:/var/run/mysqld/mysqld.sock \
-B $PWD/conf.txt:$PASA_HOME/pasa_conf/conf.txt \
-B $PWD/cdna_alignment_mysqlschema:$PASA_HOME/schema/cdna_alignment_mysqlschema \
${pasa_container} \
$PASA_HOME/Launch_PASA_pipeline.pl \
-c annotCompare.config \
--annot_compare \
-L \
--annots ptua_pasa.gene_structures_post_PASA_updates.2550175.gff3 \
-g ${genome_fasta} \
-t all.transcripts.fasta.clean \
--CPU ${threads}

6 PASA OUTPUTS

6.1 Generate checksums

cd "${pasa_output_dir}"

md5sum ptua_pasa.gene_structures_post_PASA_updates.3761026.gff3 | tee ptua_pasa.gene_structures_post_PASA_updates.3761026.gff3.md5

md5sum ptua_pasa.gene_structures_post_PASA_updates.3761026.bed | tee ptua_pasa.gene_structures_post_PASA_updates.3761026.bed.md5

3f86269b7aec49a2f1aabbfe0786f31a  ptua_pasa.gene_structures_post_PASA_updates.3761026.gff3
d258401d384efb01543b030a72ec77ac  ptua_pasa.gene_structures_post_PASA_updates.3761026.bed

6.2 Rename outputs

cd "${pasa_output_dir}"
cp ptua_pasa.gene_structures_post_PASA_updates.3761026.gff3 "${pasa_gff}"

cp ptua_pasa.gene_structures_post_PASA_updates.3761026.bed "${pasa_bed}"

md5sum "${pasa_gff}" | tee "${pasa_gff}".md5
md5sum "${pasa_bed}" | tee "${pasa_bed}".md5

3f86269b7aec49a2f1aabbfe0786f31a  ptua-PASA.gff3
d258401d384efb01543b030a72ec77ac  ptua-PASA.bed

6.3 GFF3 Preview

head -n 50 "${pasa_output_dir}"/"${pasa_gff}"

# PASA_UPDATE: mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1, single gene model update, valid-1, status:[pasa:asmbl_433257,status:3], valid-1
Pocillopora_meandrina_HIv1___xfSc0000885    .   gene    25319   26417   .   +   .   ID=gene-Pocillopora_meandrina_HIv1___TS.g4079.t1;Name=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1
Pocillopora_meandrina_HIv1___xfSc0000885    .   mRNA    25319   26417   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1;Parent=gene-Pocillopora_meandrina_HIv1___TS.g4079.t1;Name=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1
Pocillopora_meandrina_HIv1___xfSc0000885    .   exon    25319   25469   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1.exon1;Parent=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1
Pocillopora_meandrina_HIv1___xfSc0000885    .   CDS 25319   25469   .   +   0   ID=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1.cds.1;Parent=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1
Pocillopora_meandrina_HIv1___xfSc0000885    .   exon    26266   26417   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1.exon2;Parent=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1
Pocillopora_meandrina_HIv1___xfSc0000885    .   CDS 26266   26417   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1.cds.2;Parent=mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1


#PROT mrna-Pocillopora_meandrina_HIv1___TS.g4079.t1 gene-Pocillopora_meandrina_HIv1___TS.g4079.t1   MAADSDSQKRKRLAGAHYKNTELPQEDEPLTLPKHYIGEQLAGEFGRVDTGPVLIQRKDIRLKKTPKKIGSDWRPSRTKSLQVIAIRAAKLRGGKPDEHG*

# PASA_UPDATE: mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1, single gene model update, valid-1, status:[pasa:asmbl_432824,status:3], valid-1
Pocillopora_meandrina_HIv1___xfSc0000812    .   gene    9924    23544   .   +   .   ID=gene-Pocillopora_meandrina_HIv1___RNAseq.g77.t1;Name=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   mRNA    9924    23544   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1;Parent=gene-Pocillopora_meandrina_HIv1___RNAseq.g77.t1;Name=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    9924    9945    .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon1;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 9924    9945    .   +   0   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.1;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    10159   10230   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon2;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 10159   10230   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.2;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    11013   11102   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon3;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 11013   11102   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.3;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    13397   13465   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon4;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 13397   13465   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.4;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    15786   15809   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon5;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 15786   15809   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.5;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    16488   16556   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon6;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 16488   16556   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.6;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    17143   17391   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon7;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 17143   17391   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.7;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    17592   19513   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon8;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 17592   19513   .   +   2   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.8;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    21811   23544   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.exon9;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 21811   22104   .   +   0   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.cds.9;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   three_prime_UTR 22105   23544   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1.utr3p1;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1


#PROT mrna-Pocillopora_meandrina_HIv1___RNAseq.g77.t1 gene-Pocillopora_meandrina_HIv1___RNAseq.g77.t1   METAKWEGTSSLTVVPTTGHGRELEESRGGTDTVPLLVSPTSEPEEELEPGELRMELPRMKDISAEEESCKSDDPSENKRRRLADLAEQLITDNSTGEESFKSDDPSENKRRRLAEPLNVEKCQEQLKSYYNTFSKVKIIPWDDSSSIQIDEIYTPLSWVRDHRKPSGVTQEELEDYTDMFKEKPTRMLVYGRPGIGKNKVLLILDGYDEYSFAEEHSPILEIWKGELLRDCHVIVTTRQLKCDELRGPSHVQLEIQGFKSRERKETFARKFMAGEEDLDEFNLYLEEKDLYDMAEIPLLLLMLCSLWKEKRHEGLPKSRADIFTQFIQTMLDHKGGSHQSMPFQKVTSTEAREDLSNLGKAAFEALLQDRLYVRCIKLPGNISRSLEKLSEVGLFQIVNLTSLNPERGAYFIHKSVQEFLAAWHIKEEVLSNKGESTLSLSKVESFEEIVKMKEVLKFACELSTEAACAVFRHVGSVGRKESVSEFDFIELLLEDEELPVNEEVYHELIWHSYFCCSAEKRRDLCSVFPSCTGGGFLYLDSNRVNITANEHLLKSGMIPDFIFFPDYENSSEKSYRDLITVAEDTNAVFLSRSGEKKAADVLKKFPRRPMDEFFLKRERKIVVYVNQIRKGRNVSTFPTEMLRELISPTAESTQVTRLVDPLNEHDRETASSFTQNTDSITGPTPQSLSRVKQIDIVGIERQEIKMLADFLPLFTALRRIDIYGEPFEIIAAQLTETLVSRIIFNDRLHTLVLANINLTAKPAAVIARSLHQANGLQRLGLSWNPLGEGVSVLIQHLSRVPHLEWLWLSEVKMTKQQVNDLSAAVRQSNISWLETDYHDCKGNVKPEEEWPTDEYWSDYWWESEEESDPGSVTDSGEEEEPGSVTNSGDEGDPGSVTDSSDQEDPGLVTDSGGEEDPGSVTDSGDKEEPGSCLET*

# PASA_UPDATE: mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1, single gene model update, valid-1, status:[pasa:asmbl_432858,status:12], valid-1
# PASA_UPDATE: mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.1.6992b91d, single gene model update, valid-1, status:[pasa:asmbl_432855,status:12], valid-1
Pocillopora_meandrina_HIv1___xfSc0000812    .   gene    34067   37241   .   +   .   ID=gene-Pocillopora_meandrina_HIv1___RNAseq.g78.t1;Name=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   mRNA    34073   35237   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1;Parent=gene-Pocillopora_meandrina_HIv1___RNAseq.g78.t1;Name=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    34073   34168   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.exon1;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 34073   34168   .   +   0   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.cds.1;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    34483   34488   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.exon2;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 34483   34488   .   +   0   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.cds.2;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    34894   34988   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.exon3;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 34894   34988   .   +   0   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.cds.3;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   exon    35222   35237   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.exon4;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   CDS 35222   35237   .   +   1   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.cds.4;Parent=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1
Pocillopora_meandrina_HIv1___xfSc0000812    .   mRNA    34067   37241   .   +   .   ID=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1.1.6992b91d;Parent=gene-Pocillopora_meandrina_HIv1___RNAseq.g78.t1;Name=mrna-Pocillopora_meandrina_HIv1___RNAseq.g78.t1

6.4 GFF Comparisons

printf '%s\n\n' "Original GFF feature counts:"
awk '!/^#/ && !/^[[:space:]]*$/ && NF > 0 && $3 != "" {print $3}' ${genome_gff} \
| sort | uniq -c | sort -rn | awk '{print $2, $1}'

echo ""
echo "${line}"
echo ""

printf "%s\n\n" "Updated GFF feature counts:"
awk -F "\t" '!/^#/ && !/^[[:space:]]*$/ && NF > 0 && $3 != "" {print $3}' "${pasa_output_dir}"/"${pasa_gff}" \
| sort | uniq -c | sort -rn | awk '{print $2, $1}'

Original GFF feature counts:

exon 208535
CDS 208535
transcript 31840

-----------------------------------------------

Updated GFF feature counts:

exon 351589
CDS 341675
mRNA 43292
gene 32019
five_prime_UTR 22150
three_prime_UTR 21812

7 EXTRACT PROTEINS TO FASTA

cd "${pasa_output_dir}"

awk '/^#PROT / {print ">" $2 "." $3 "\n" $4}' "${pasa_gff}" > ptua-proteins-PASA.fasta

printf "%s\n\n" "Original protein counts:"
grep --count "^#PROT" "${pasa_gff}"

echo ""
echo "${line}"
echo ""

printf "%s\n\n" "Extracted protein counts:"
grep --count "^>" ptua-proteins-PASA.fasta

# Create FastA Index
${samtools} faidx ptua-proteins-PASA.fasta

Original protein counts:

43292

-----------------------------------------------

Extracted protein counts:

43292

7.1 Checksums

cd "${pasa_output_dir}"
md5sum ptua-proteins-PASA.fasta | tee ptua-proteins-PASA.fasta.md5

d41376d2ac0a8f41b0bbe333ed8cf08b  ptua-proteins-PASA.fasta

PASA Web Portal - Configuration on Gannet Using Synology Container Manager

Sam White — Fri, 27 Feb 2026 08:00:00 GMT

INTRO

The PASA pipeline (GitHub) was used to generate transcriptome annotations using the RNA-seq data for the three coral species in the `timeseries_molecular project (GitHub) project. The PASA web portal provides a user-friendly interface to explore and visualize the results of the PASA pipeline, including gene models, alignments, and annotations.

Below, I detail the steps I took to configure the PASA web portal on Gannet using the Synology Container Manager. This includes building a custom Docker image with the necessary Perl modules, transferring it to the Synology, and setting up the container to run the PASA web portal.

METHODS

Note

The example below uses the peve_pasa database, but you can replace it with your database name as needed.

Build and Transfer Docker Image

Note

This doesn’t have to be done on a local computer; you can build the image directly on the Synology if you have the necessary tools and permissions. However, building locally and transferring can be faster and more convenient in some cases.

Build the image (if not already done)

cd /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/PASA
sudo docker build -t pasapipeline-fixed:latest -f Dockerfile.pasaweb .

Here’s what’s in Dockerfile.pasaweb:

$ cat Dockerfile.pasaweb

FROM pasapipeline/pasapipeline:latest

# Install missing Perl modules required for PasaWeb
RUN cpan -i CGI GD::Graph

# Keep the original entrypoint

Save the image to a tar file.

sudo docker save pasapipeline-fixed:latest \
> /home/sam/pasapipeline-fixed.tar

Transfer to Synology (replace with your Synology details).

rsync -avP pasapipeline-fixed.tar \
gannet:/volume2/docker/PASA_web_portal/

Synology Configuration Steps

Command Line Setup

On Synology, via command line:

Load the pre-built image


cd /volume2/docker/PASA_web_portal
sudo docker load -i pasapipeline-fixed.tar

Grant MySQL permissions (if not already done):

mysql -u root -p<password>

GRANT SELECT ON peve_pasa.* TO 'pasa_access'@'localhost';
GRANT SELECT ON apul_pasa.* TO 'pasa_access'@'localhost';
GRANT SELECT ON ptua_pasa.* TO 'pasa_access'@'localhost';
FLUSH PRIVILEGES;

Create/import databases (if not already done):

mysql -u root -p -h127.0.0.1 -P3307 <<'EOF'
CREATE DATABASE IF NOT EXISTS peve_pasa;
EOF

Fix collation issue in the SQL dump file (if needed):

sed -i 's/utf8mb4_0900_ai_ci/utf8mb4_general_ci/g' '/volume2/web/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/PASA/peve_pasa_backup.sql'

Load the SQL dump into the database:

mysql -u root -p<password> -h127.0.0.1 -P3307 peve_pasa \
  < /volume2/web/gitrepos/urol-e5/timeseries_molecular/E-Peve/output/00.30-E-Peve-transcriptome-assembly-Trinity/PASA/peve_pasa_backup.sql

Web Interface

Then, access the Synology via the web interface and set up the container.

In the Container Manager app, create a new container with these settings:

Go to the “Image” tab You should see pasapipeline-fixed:latest in the list of images.

Click on the image, then click “Launch” or “Run”

This opens the container creation wizard where you configure: - Container name - Port mappings - Volume mounts - Environment variables - xecution Command: bash -c "cp /pasa_data/conf.txt /usr/local/src/PASApipeline/pasa_conf/conf.txt && /usr/local/src/PASApipeline/run_PasaWeb.pl 9000" - Port Settings:

- Local Port: 9000
- Container Port: 9000
- Type: TCP
- Volume Mappings:

Host: /volume1/path/to/PASA/directory
Container: /pasa_data
Network:
Use bridge mode
Working Directory:

Access

Important

http is used instead of https because the PASA web portal does not support SSL/TLS. Ensure that your network is secure and consider using a VPN if accessing remotely.

http://gannet.fish.washington.edu:9000 Enter database name when prompted (e.g., peve_pasa)

qPCR Analysis - M.gigas PolyIC Data from Valentinas Project

Sam White — Mon, 02 Feb 2026 08:00:00 GMT

INTRO

This is an analysis of qPCR data from Magallana gigas (Pacific oyster) samples treated with PolyIC (a synthetic analog of double-stranded RNA that mimics viral infection) and subjected to different stress types (Temperature, Mechanical, Control). The goal was to assess gene expression changes in response to these treatments.

The qPCR data was generated by me on January 27-29, 2026, as part of Valentina’s project on immune responses in oysters. Eight genes were targeted: ATP Synthase, Citrate Synthase, DNMT1, HSP70, HSP90, cGAS, and VIPERIN. GAPDH was used as the normalizing gene. Notebook entries documenting the qPCR runs and data exports can be found in the following posts:

I also managed to track down the qPCRs run on the RNA I isolated on isolated on 20250804. After RNA isolation, they were passed to Valentina for reverse transcription and qPCR. I believe those qPCRs were run/analyzed by Aakriti on 20250814 (Notebook entry). I found the files here:

Important

I have not taken the time to evaluate the qPCRs run by Aakriti in detail. I’ve relied on the processing below to filter out samples with poor technical replicates. Maybe these can be revisited at a later date to see if any samples need to be re-run.

Code

qPCR - M.gigas Valentina PolyIC VIPERIN

Sam White — Thu, 29 Jan 2026 08:00:00 GMT

INTRO

Continuing work with Valentina’s PolyIC project after creating cDNA on 20251019 (Notebook entry), it was time to run the qPCRs.

This notebook describes how the qPCRs were run and links to the various output files. It also provides a brief overview of each primer set’s amplification profiles. This notebook does not have any analysis. This will be performed later.

MATERIALS & METHODS

The following primer sets were run:

SR ID	Primer Name
1829	Cg_VIPERIN_F
1828	Cg_VIPERIN_R

All samples were run in triplicate, on low-profile, white 96-well plates (USA Scientific) sealed with TemPlate TR Select Optical Film (USA Scientific) in a CFX Connect (Bio-Rad) or CFX96 (Bio-Rad) real-time thermalcycler. All reactions consisted of the following:

Component	Stock Concentration	Volume (uL)
cDNA	NA	1
SsoAdvanced Universal SYBR Green Supermix (BioRad)	2x	10
P_F	10uM	0.5
P_R	10uM	0.5
H₂O	NA	8
TOTAL		20

Master mixes were distributed across three plates for each primer set and included no template controls (NTCs).

For cycling parameters, plate layouts, etc. see the RESULTS section below.

RESULTS

Important

There were a number of samples which ran out or cDNA volume during the qPCR setup. So, not all samples have three replicates. These are noted in the plate layouts in the PDF reports linked below.

These missing samples were removed from the qPCR analysis, so that’s why those wells are missing the Cq results files.

Summary

VIPERIN: Amplification and melt plots across all three plates.

Files

Tip

report-*.pdf: qPCR Reports. Contains plate layouts, cycling params, amp/melt plots, etc.
*Cq_Results.csv: Cycle quantity (Cq) data.
*.pcrd: Source qPCR data file. Requires CFX Maestro (Bio-Rad) software to open.