Mapping against an inverted reference chromosome to check for an inversion?
0
0
Entering edit mode
6.2 years ago
orlando.wong ▴ 60

Hello Biostars,

I mapped my mate-pair data against inverted chromosome 9 sequence in an attempt to find a few large inversions (1 million base pairs+) that may have resulted from equally large deletions.

Steps I took:

1) Downloaded the human chr9.fa from UCSC Genome Browser.

2) reversed the sequence with a C program that reads the file backwards and writes into a new file.

3) I called the new file chr9inv.fa. I changed the beginning of the file from >chr9 to >chr9inv

4) I then used this file as a reference for BWA-MEM alignment on Galaxy servers.

Very few reads were aligned, and the ones that were were not properly aligned. I am not sure why.

Does BWA algorithm have issues when aligning with very large deletions in the genome? I may have done some error that I am not aware of. Any ideas?

bwa Bowtie mapping NGS inversion • 1.8k views
ADD COMMENT
2
Entering edit mode

Err, think about what you did biologically...

ADD REPLY
0
Entering edit mode

I drew it out biologically:

Sample Sequence

5' -A T C C G - 3'

3' -T A G G C - 5'

Would the inverted sequence be like -

(rotated 180 degrees)

5' - C G G A T - 3'

3' - G C C T A - 5'

or (flipped over)

5' - G C C T A - 3'

3' - C G G A T - 5'

ADD REPLY
1
Entering edit mode

The former.

ADD REPLY
0
Entering edit mode

I will adjust to reverse-complement the reads. Thank you!

ADD REPLY
0
Entering edit mode

what is the purpose of the step (2) ??

ADD REPLY
0
Entering edit mode

I was thinking that if my case had an inversion, then there will be a few reads that will map successfully against an inverted chromosome 9. What do you think?

ADD REPLY
1
Entering edit mode

BWA mem will align reads across inversions without you needing to do that (further, you're just swapping the strands and the alignment is against both of them anyway). Look for supplemental alignments, which would indicate possible points of inversion.

ADD REPLY
1
Entering edit mode

what Devon said. Furthermore, for a mate-pair (not paired-end), I think you'll have to reverse-complement one the two fastq files.

ADD REPLY

Login before adding your answer.

Traffic: 1464 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6