Extracting specific regions(ex, chr1:100-200 of hg38) from whole genome .vg and .gam file

Question

[hd00ljy3]

I want to extract specific regions like chr1:100-200 of hg38 from whole genome .vg and .gam file

---

I built a .vg from structural variation VCF file and I want to examine if the records in VCF are well represented in the .vg file.

What I want to do is as follows.

I have an Insertion(INS) at chr1:100 on hg38 I want to get nodes in chr1:50-200 along with nodes representing the Insertion.

But since I can not find any labels(variant IDs or chromosomal coordinates) on the graph when I converted .vg to .dot, I was not able to extract specific regions or nodes adjacent to specific SVs.

I looked through the wiki and github issues but I was not able to find solutions.

Could you help me on this?

Answer 1

[glenn.hickey]

You can extract regions from a GAM with vg chunk but it's clunky.

Index the GAM (very slow):

vg gamsort aln.gam -i aln.sorted.gam.gai > aln.sorted.gam

Extract the chunk:

vg chunk -x graph.vg -a aln.sorted.gam -g -p chr1:100-200 -c 10 > chunk.vg

The -c 10 specifies how far from the reference path to search from to cover, say, insertions.