About Trim The Exome-Seq

Question

[camelbbs]

HI I want to ask if my exome-seq reads need to be trimmed. The fastqc quality like this: http://postimg.org/image/hrhyss5qt/

If I don't trim the last 10 bp, how it will affect my variants calling.

Thanks in advance.

[Rm]

I would rather trim at 90bp: little conservative but it helps to get quality calls

[Sean Davis]

I think trimming by base quality is probably a better approach than trimming at a certain base. Based on the FASTQC results, there are probably many high-quality bases in many reads that extend beyond 90bp. Along the same lines, there are likely reads with low quality bases occurring long before 90bp.

Answer 1

[Devon Ryan]

Most of the variant callers themselves take base quality into account when making calls, so I wouldn't be overly concerned about that. However, your mapping will also be affected by the low quality bases, which will decrease the quality of alignments fed to the variant caller and possibly increase the false positive and false negative calls. In general, you'll be better off quality trimming your data with trimmomatic or trim_galore (or something like that). Always remember the saying, "Garbage in, garbage out".