different reads length samples in same experiment

[eleven]

Hello! everyone.

I am working on the Hiseq PE (DNA) data, and some samples have 100bp PE and other samples have 90bp PE. So my question is if we can use those data for analysis in one experiment and same methods without any processing (after QC), or we can trim the different length to the same length such as 70bp, then analysis them using same methods. So which method is better?

Thank you!

[mastal]

different reads length samples in same experiment

It depends what you are going to do with your data, and what software you are going to use to analyse it.

Some programs might want the reads to be all the same length.

Some applications, such as de novo assembly, might work better if the reads are trimmed to remove adapter sequences and poor quality bases.

[westerman]

Going along with Mastal's comments, since you have genomic PE data then it is likely that you with either be doing mapping to a known genome or a de-novo assembly. In either case most programs will not care about differing lengths. Perhaps obviously once you remove adapter sequences then you will have different read lengths. Since most programs work with trimmed reads (although this step is often not needed for known genome mapping) then the programs are capable of working with different lengths.

What I find is that most programs are not capable of working with paired end and single end data in the same file nor, when working with PE data only, having the pairs out of order.

Also I'd say that at least half of the programs I use are not capable of mixing up different types of reads; e.g., handling 454 reads along with Illumina hiSeq reads.

If you want to tell us explicitly what you want to do and/or the programs you are going to use then we can give you more specific help.