Purpose

This analysis of dataset 1 (Proestou et al. 2023) is part of the post-analysis data integration approach, where RNAseq datasets are processed independently and the differential abundance pipeline is run on each to identify DEGs. The lists of DEGs will then be compared across studies.

This study excluded controls from their main analysis and focused on family differences in dosed animals. Because there isn’t currently a way to account for controls in multi-factor model, I wondered if I could first run the differential abundance pipeline on infection vs. control, then on breed to look for family differences.

Methods

Step 1: contrast controls vs. injected

input data:

id,variable,reference,target
trait_dose,condition,control,injected

run pipeline:

screen -S dfab_1
salloc -A srlab -p cpu-g2-mem2x -N 1 -c 1 --mem=16GB --time=12:00:00
mamba activate nextflow 

nextflow run nf-core/differentialabundance \
-c /gscratch/srlab/strigg/bin/uw_hyak_srlab.config \
-resume \
--study_name dataset_1 \
--input study1_samplesheet.csv \
--contrasts study1_contrasts.csv \
--outdir /gscratch/scrubbed/strigg/analyses/20250730_diffabund \
--matrix salmon.merged.gene_counts_length_scaled_dataset1.tsv \
--feature_type gene \
--features_gtf_feature_type gene \
--filtering_min_abundance 10 \
--filtering_min_proportion 0.75 \
--gtf /gscratch/srlab/strigg/GENOMES/GCF_002022765.2_C_virginica-3.0_genomic_noEmptyGeneIDs.gtf


#this attempt didn't have the threshold so I reran as the code above
	
nextflow run nf-core/differentialabundance \
-c /gscratch/srlab/strigg/bin/uw_hyak_srlab.config \
-resume \
--study_name dataset_1 \
--input study1_samplesheet.csv \
--contrasts study1_contrasts.csv \
--outdir /gscratch/scrubbed/strigg/analyses/20250730_diffabund/ds_1 \
--matrix salmon.merged.gene_counts_length_scaled_dataset1.tsv \
--feature_type gene \
--features_gtf_feature_type gene \
--gtf /gscratch/srlab/strigg/GENOMES/GCF_002022765.2_C_virginica-3.0_genomic_noEmptyGeneIDs.gtf

report: dataset_1.html

output: 20250730_diffabund/ds_1

Step 2: contrast families of injected animals using the DEGs from the step 1 contrast

input files:

id,variable,reference,target
trait_dose,trait,sensitive,resistant

#filter the gene counts matrix for DEGs from step 1
awk -F "\t" 'NR==FNR{a[$1]=$1;next}$1 in a{print $0}' tables/differential/trait_dose.deseq2.results_filtered.tsv salmon.merged.gene_counts_length_scaled_dataset1.tsv > salmon.merged.gene_counts_length_scaled_dataset1_ctrlFilt.tsv
  • samplesheet:
    • wget https://gannet.fish.washington.edu/metacarcinus/USDA_MetaOmics/Cvirg_RNAseq/20250702_diffabund/study1_samplesheet.csv

Run pipeline on filtered data:

nextflow run nf-core/differentialabundance \
-c /gscratch/srlab/strigg/bin/uw_hyak_srlab.config \
-resume --study_name dataset_1 \
--input ctrl_filt/study1_samplesheet.csv \
--contrasts ctrl_filt/study1_contrasts.csv \
--outdir /gscratch/scrubbed/strigg/analyses/20250730_diffabund/ds_1/ctrl_filt \
--matrix ctrl_filt/salmon.merged.gene_counts_length_scaled_dataset1_ctrlFilt.tsv \
--feature_type gene --features_gtf_feature_type gene --filtering_min_abundance 10 \
--filtering_min_proportion 0.75 --gtf /gscratch/srlab/strigg/GENOMES/GCF_002022765.2_C_virginica-3.0_genomic_noEmptyGeneIDs.gtf

Copy data to Gannet

rsync --progress --verbose --archive 20250730_diffabund shellytrigg@gannet.fish.washington.edu:/volume2/web/metacarcinus/USDA_MetaOmics/Cvirg_RNAseq

Results

PCA of DEseq2 DEGs

The PCA shows a good separation of the sensitive vs. tolerant animals using the 26 significant DEGs identified in step 2, which makes the results believable.