technical replicates

[animesh]

Description of the bug

not sure how to handle technical replicates in contrast file? i have a samples.csv like one presented at https://nf-co.re/differentialabundance/1.4.0/docs/usage but with one more column (tech) representing technical replicate

sample|fastq_1|fastq_2|condition|replicate|tech|batch
CONTROL_REP1T1|AEG588A1_S1_L002_R1_001.fastq.gz|AEG588A1_S1_L002_R2_001.fastq.gz|control|1|1|A
CONTROL_REP1T2|AEG588A1_S1_L003_R1_001.fastq.gz|AEG588A1_S1_L003_R2_001.fastq.gz|control|1|2|B
CONTROL_REP1T3|AEG588A1_S1_L004_R1_001.fastq.gz|AEG588A1_S1_L004_R2_001.fastq.gz|control|1|3|A
...
TROL_REPNT1|AEG588AN_S1_L002_R1_001.fastq.gz|AEG588A1_S1_L002_R2_001.fastq.gz|trol|1|1|A
TROL_REPNT2|AEG588AN_S1_L003_R1_001.fastq.gz|AEG588A1_S1_L003_R2_001.fastq.gz|trol|1|2|B
TROL_REPNT3|AEG588AN_S1_L004_R1_001.fastq.gz|AEG588A1_S1_L004_R2_001.fastq.gz|trol|1|3|A

but i am not sure what should be the best way to present this information in contrasts.csv?

Command used and terminal output

No response

Relevant files

No response

System information

No response

[pinin4fjords]

Assuming this is RNA-seq, we would normally expect technical replicates (assumed to mean multiple sequencing runs of the same biological sample) to have been collapsed in the upstream analysis- see https://nf-co.re/rnaseq/3.13.2#usage.

Do you have reason to handle individual tech reps separately?

The DESeq2 module in nf-core doesn't currently have capability for collapsing the tech reps there, so that would need addressing before we could handle them as part of this workflow.

[animesh]

Thanks @pinin4fjords for the response. I just wanted to see how well the tech-reps are clustering and they were clustering quite well 👍🏽 So i tried to merge the counts by summing and presented the sum of tech-reps as bio-rep in contrast for RNA-seq pipeline, which seems to have worked 🤞 Probably this process can be part of pipeline itself?

[pinin4fjords]

Thanks- we'll consider taking matrices with unmerged tech reps in future development.