Gene alignment with IGV
1
1
Entering edit mode
2.8 years ago
flappix ▴ 30

Hi,
I have loaded a genome and a gtf file into IGV to see the intron/exon structure of a gene. Now I want to align a sequence against this gene and see which parts of the sequence are falling into introns and exons of the gene.

Which program would be best to create the alignment in way IGV can read it? Which file format would be best?

When google for IGV and alignment I always come up with bam format. I think this is not very practical to use for my aim.

I included a picture of my imagination how it should look like (the green track I draw by myself, this should be the alignment file).

I would be very happy if someone have a hint for me :)

edit: In case anyone is interested: I started to write my own viewer for this purpose:
https://github.com/flappix/alignment-viewer

igv alignment • 1.9k views
1
Entering edit mode

First, you must use alignment tools; bwa, bowtie or something like that if you want to do this. However, I do not know what kind of data you have, like transcriptome, or genome data ? The tools you need to use that It depends the data you have.

My advice is that; You can extract the region you want in the genome using script. Then, make it reference and align sequences you have against it. I think, It must work. Then, you should upload all the files you have, fasta, gtf, and bam, to IGV.

0
Entering edit mode

Thanks for your suggestion. I tried this but I was not able to get a reasonable alignment with bwa. I think these programs are designed to map short reads against a large reference genome and not to align two (somewhat) similar genes.

0
Entering edit mode

Have you tried to use blat (or GMAP) standalone?

1
Entering edit mode
2.8 years ago
ATpoint 71k

IGV offers a BLAT function via Tools => BLAT where you can paste and then align your sequence with. Not sure if this will 100% be what you need but since it is super simple I'd try it first. It outputs a window with all significant hits.

In the example below I BLATed a random part of the ETS2 coding sequence (so no introns, only a subset of the coding exons / cDNA). BLAT/IGV finds the origin in the mm10 genone and creates a track with intron/exon structure. Note that the introns were automatically derived by IGV. Quite handy feature I think.

0
Entering edit mode

Thanks for your suggestion. Unfortunately one can only use this feature if the genome is available in the UCSC genome browser which is not the case (or one have to setup its own Blat server, but that's a bit of an overkill). Nevertheless it's good to know this feature exists in case I face this problem with a another genome.