Very low Bisulfite seq mapping efficiency (~0%) for paired end.

Question

[cjgunase]

Hello,

I have bisulfite converted fastq files (WGBS) from paired-end reads. when I used bismark using the default setting for paired-end data mapping efficiency is almost zero. Then I tried using read1 and the mapping efficiency is 85%.

I have put my mapping report in this link. If anyone can give suggestions to improve the paired-end mapping efficiency I would be really grateful.

https://drive.google.com/drive/folders/0B8CAUxn-nmdFMThJWkQzX1MwNUE?usp=sharing

Thank you

[Devon Ryan]

From a cursory look at the metrics, it looks like there's an issue with read #2. Have you run FastQC on it?

[cjgunase]

yes. FastQC looks perfect on both files. found possible reasons in http://seqanswers.com/forums/showthread.php?t=40496

we believe it is over clustering which can happen in read2.

Not sure how to prepare a report so we can request re-sequencing. If you can suggest some tools and packages to do more investigation, that would be great.

Thank you

[Devon Ryan]

Actually, your files are in a different order, so the first read in the #1 file is not the mate of the first read in the #2 file.

[cjgunase]

Thanks Devon,

Can you kindly point out how did you find this so I can narrow down where the problem originated? Is there anything to correct this?

Thank you

[Devon Ryan]

I think BBMap has a tool to resync paired-end files.

[cjgunase]

Thankx Devon, I tried BBMap and this script worked even better. now its mapping in paired-end with ~85% https://github.com/enormandeau/Scripts/blob/master/fastqCombinePairedEnd.py

Thank you for your effort to help solve this issue.

[jihed.chouaref]

Thanks for the script, I am encountering the same issue right now! I hope it helps :)

[cjgunase]

I did a comparison as you said. Thank you for the idea. I am going to reorder the fastq files.

Not aligning sample(1F52S)

[gunasekara@sphere fq0050]$ head -1 split.10m.GTEX-1F52S-3026-SM-D5A5F_1.fq @E00591:56:H2LTKCCXY:3:1102:31527:12683 1:N:0

[gunasekara@sphere fq0050]$ head -1 split.10m.GTEX-1F52S-3026-SM-D5A5F_2.fq @E00591:56:H2LTKCCXY:3:1102:20841:12736 2:N:0

Aligning sample (14PQA)

[gunasekara@sphere 1F52S_test]$ head -1 split.10m.GTEX-14PQA-0926-SM-D5A5H_1.fq0000 @E00591:57:H2LNVCCXY:6:1101:23937:4069 1:N:0

[gunasekara@sphere 1F52S_test]$ head -1 split.10m.GTEX-14PQA-0926-SM-D5A5H_2.fq0000 @E00591:57:H2LNVCCXY:6:1101:23937:4069 2:N:0