Question: illumina PE quality control - NGS: QC & Manipulation
gravatar for sumudu_rangika
2.2 years ago by
sumudu_rangika0 wrote:


I have a PE illumina miseq data set (separate forward-R1 & reverse-R2) of a WGS of a parasite. I'm using a local instance of galaxy. I have run FASTQC on sequencer given data sets and decided to do some quality control since quality drops towards the end of reads.

I have few doubts to clear and appreciate if galaxy team or anyone in the forum could help me to solve them.

1) According to FASTQC I decided to quality trim reads using FASTQ quality trimmer (sliding window=5, step=2, avg quality=20). Since quality degrades towards 3' end should I trim 3' end only. Is this Ok?

2) I have seen suggestions to perform quality control for PE data sets (R1 & R2) together & not separately. Isn't it correct to perform QC on R1 & R2 separately?

3) If so, I have to use another tool instead of FASTQ quality trimmer. Would it be a good option to use trimmomatic or Trim galore?

4) Later I used Filter FASTQ to filter reads <70bp; If I use R1 dataset only should I use yes/no to "This is paired end data" option? Can I select multiple data sets option in filter FASTQ to choose both R1&R2 together?

Thanks in advance, Sumudu

ADD COMMENTlink modified 2.2 years ago by Jennifer Hillman Jackson25k • written 2.2 years ago by sumudu_rangika0
gravatar for Devon Ryan
2.2 years ago by
Devon Ryan1.9k
Devon Ryan1.9k wrote:
  1. A quality of 20 is overkill for many use cases (unless you're trying to do assembly maybe, I can't give advice there).
  2. Only ever use trimmomatic, or trim galore, or a similar program with paired-end datasets. I would also suggest simply removing fastq quality trimmer, since inevitably having it there will lead someone to try to use it with paired-end data and get completely weird results.
  3. See above
  4. Both trimmomatic and trim galore have options to filter out reads below a given size, just use them. Regarding the "Filter Fastq" tool, it looks like the "paired end" option somehow signals that the paired-ends have been merged together, which wouldn't be the case for you.
ADD COMMENTlink written 2.2 years ago by Devon Ryan1.9k

Thank you very much Devon.

ADD REPLYlink written 2.2 years ago by sumudu_rangika0
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 172 users visited in the last hour