Tweet
Hi all,
I already checked the forum but did now find the answers.
I am about to find the germline SVs in sequenced trios. Therefore, I thought that it is a good idea to ask for a few tips on how it should be done.
I have read a few papers and found SV callers. Most of the materials treat about somatic mutations, which are not of my interest.
QC and mapping are well understood by me. The problem lies in finding the inherited mutations.
Should I map reads from each member in the trio to the reference genome and from these SAM/BAM files looking for germline SVs?
And the thing is, how do i know, that this is a germline one? Should I run the SV caller on each BAM file to find the SVs, but how should I then approach the characterisic of my result (merging these VCFs somehow, or some scripting to know which SVs are contained in son/daughter which are not - in other words looking for the mendelian inheritance patterns).
Finally I would expect to merge the blocks in haplotypes, but that's a distant future.
Another thing is to use a tool for pooled SV calling, but I have no idea what are the principles behind their work.
My description might be a bit blury, but it is a beginning, and I do not have a clear view of how to approach the problem.
So with general ideas, if you were about to pull out this, how would you approach the problem?
I already checked the forum but did now find the answers.
I am about to find the germline SVs in sequenced trios. Therefore, I thought that it is a good idea to ask for a few tips on how it should be done.
I have read a few papers and found SV callers. Most of the materials treat about somatic mutations, which are not of my interest.
QC and mapping are well understood by me. The problem lies in finding the inherited mutations.
Should I map reads from each member in the trio to the reference genome and from these SAM/BAM files looking for germline SVs?
And the thing is, how do i know, that this is a germline one? Should I run the SV caller on each BAM file to find the SVs, but how should I then approach the characterisic of my result (merging these VCFs somehow, or some scripting to know which SVs are contained in son/daughter which are not - in other words looking for the mendelian inheritance patterns).
Finally I would expect to merge the blocks in haplotypes, but that's a distant future.
Another thing is to use a tool for pooled SV calling, but I have no idea what are the principles behind their work.
My description might be a bit blury, but it is a beginning, and I do not have a clear view of how to approach the problem.
So with general ideas, if you were about to pull out this, how would you approach the problem?