GC content for RNA-Seq data

Question

[sukesh1411]

Hi

I am new to the rna-seq data analysis. I have rna-seq data of leaf. I would like to perform transcriptome assembly using trinity. Before doing it, i removed the adapters and low quality bases. When i check the quality of reads in FASTQC it shows error for per base sequence content, per base GC content and sequence duplication levels. I dont know how to normalize this data before going to perform denovo assembly

Fastqc results

Thanks

[WouterDeCoster]

FastQC is probably overreacting again, doesn't look like something to worry about in my opinion.

Answer 1

[GenoMax]

Don't do anything. That is very characteristic signature of RNAseq data in FastQC. If you were over cautious in removing low quality bases then dial back on that as well. You probably need to filter at Q20 or below if you truly have bad quality data.

[biofalconch]

Hi genomax, just wondering, is it necessary to account for GC content when doing the differential expression analysis? Based on what it says on this article

[WouterDeCoster]

In general, no. it's not. Because you are comparing the expression of gene X with gene X across samples, they will all have the same GC content.

If you would compare gene X (GC=35% ) with gene Y (GC=76%) then that would matter, but that's not the scope of differential expression analysis.

[biofalconch]

Thank you for your answer Wouter :)

[russhh]

I guess it depends whether there is a GC bias between different sequencing samples