Discordant number of TE copies across samples

[juheon]

Hi,

I am looking into TE expression upon drug treatment using stranded RNA-seq of two samples.
Here are the commands I used
squire Count -c <clean_dir> -f <fetch_dir> -r 75 -n <sample fastq> -p 4 -b hg38 -s 2 squire Call -1 -2 -A treated -B control -i squire_count -o -p 4 -f pdf

The number of TE copies in "DESeq2_TE_only.txt" is not the same across samples while the number of genes in "DESeq2_RefSeq_only.txt" is the same. (To avoid the case where one TE copy/gene generates two transcripts in the opposite strand, I trimmed the transcript strand information in row names and count the number of TE copy and gene.)

• sample A: 28277 gene (from DESeq2_RefSeq_only.txt), 113262 TE copies (from DESeq2_TE_only.txt)
• sample B: 28277 gene (from DESeq2_RefSeq_only.txt), 159599 TE copies (from DESeq2_TE_only.txt)

I expect to see all TE copies in "DESeq2_TE_only.txt" but only a subset of TE copies are shown for each sample and the number of TE copies is discordant across samples. Is there a way for me to have SQuIRE output with all TE copies?

[rpg18]

Hi @juheon ,

So far as I understand, DESeq2_TE_only.txt should be the output from the DEA. It stores the results generated by DESeq2. squire Call should generate only one DESeq2_TE_only.txt per dataset. While squire Count should generate one {sample}_TEcounts.txt file per sample. Did you probably mean TEcounts.txt instead DESeq2_TE_only.txt in your post? The amount of annotated/quantified TEs in TEcounts.txt files differs across my samples too.

If you want SQuiRE output with all TE copies, you can find them in {dataset}_gene_TE_counttable.txt file generated by squire Call.

All the best,
Raquel

[juheon]

Hi, Thanks for the response. Since I ran squire for two samples separately, I have DESeq2_TE_only.txt for each sample. The reason why I ran squire separately for each sample is to see sample-specific responses upon drug treatment. In the DEseq output for TE (DESeq2_TE_only.txt), I expect to see all annotated TE locus but it was not. I checked *{dataset}_gene_TE_counttable.txt *and the number of rows is the same as DESeq2_all.txt. I think only a subset of annotated TE locus is chosen somehow for {dataset}_gene_TE_counttable.txt and that countable is used as the input for DEA. a) What is the criteria to choose a subset of annotated TE locus for {dataset}_gene_TE_counttable.txt? a) Is it valid to use a subset of annotated TE locus for DEseq in terms of multiple hypotheses correction? If separate SQUIRE runs for two samples uses different number of TE locus for DE analysis, FDR is not comparable across samples. Isn't it?

…

On Fri, Sep 25, 2020 at 12:46 PM rpg18 ***@***.***> wrote: Hi @juheon <https://github.com/juheon> , So far as I understand, DESeq2_TE_only.txt should be the output from the DEA. It stores the results generated by DESeq2. squire Call should generate only one DESeq2_TE_only.txt per dataset. While squire Count should generate one {sample}_TEcounts.txt file per sample. Did you probably mean TEcounts.txt instead DESeq2_TE_only.txt in your post? The amount of annotated/quantified TEs in TEcounts.txt files differs across my samples too. If you want SQuiRE output with all TE copies, you can find them in *{dataset}_gene_TE_counttable.txt* file generated by squire Call. All the best, Raquel — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#46 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABFCNNCIF3URRJSAQFJR23DSHTJQVANCNFSM4RXM3AUQ> .

[rpg18]

DESeq2 input is {dataset}_gene_TE_counttable.txt . This counting table include all annotated genes and a total subset of TE loci:

a) check here and here how {dataset}_gene_TE_counttable.txt is generated. SQuiRE-Call filters TE locus counts, reducing the total number of TE loci in the final counttable.txt file, what is a good point. The reason why {sample}_TEcounts.txt files have different number of TE loci is because of SQuiRE-Count approach.

b) If you run SQuiRE-Call with the whole dataset, you will obtain a unique {dataset}_gene_TE_counttable.txt, which is the input for DESeq2. That is, you obtain an equal number of entries (genes/TEloci) for all samples.

Hope this can be of help.

[mars188]

@rpg18 may be you can help with the issue I am facing while running "Count" and "Call".

I have 8 samples (two groups) that I ran Squire on and obtained individual count tables using "Count". Later, I performed differential expression of TEs with "Call". However, when I looked at the "SQuIRE_gene_TE_counttable.txt" file generated during the "Call" step, it shows different number of TE counts than the "{sample}_TEcounts.txt" file generated during the "Count" step.

I was expecting that same count values generated at Count step would be transferred over and different expression would be performed with Call? Confusingly, out of 8 samples, 3 show that TE count values match between "{sample}_TEcounts.txt" and "SQuIRE_gene_TE_counttable.txt" files but not for the rest 5 samples.

Will really appreciate your help