Pair-end sequence data preparation for Trancriptome assembly

[GillermoPonz]

Hi everyone,

I'm about to perform a transcriptome assembly with Trinity, but I have a basic question.

I have two RNA-Seq files containing 40 million Pair-End reads from Illumina Hi-Seq. After performing a FastQC I’ve observed that the first 10 bases have a bad result in the "Per Base Sequence Content" plot and I want to trim them (by trimming I mean Trimming by column, because the quality trimming didn’t fix the problem) to avoid problems during the assembly. My problem is that I don’t know if by trimming the first 10 bases of the 5’ end in both files I will mess up the sequence pair mating and completely ruin the downstream analysis.

Could anybody help me with this? I know is a basic question, but I’m a newbie and I haven’t been able to get the answer by myself.

[GenoMax]

Do not trim the data at the beginning of the read. That bias is a "known" feature of RNAseq data and it does not affect any downstream analysis.

See post #261 and #263: http://seqanswers.com/forums/showthr...t=4846&page=14

[GillermoPonz]

Thanks for your quick answer, this info has been very useful for me.

But just for curiosity, do you know if it would affect the pairing of the sequences (and in consequence the ability of Trinity to use the pair-end information) if I trim them?

[GenoMax]

As long as you use a trimming program (trimmomatic, bbduk etc) that is paired-end aware, pairing of sequences in your data would be maintained.