Identity-By-Descent Used To Narrow Down Targets From Exome Sequencing Data

Question

[michealsmith]

I know someone will apply IBD or linkage analysis to further narrow down the variant targets from exome sequencing data besides dbSNP filter or missense filter, for example:

http://bioinformatics.oxfordjournals.org/content/27/6/829.long

In this paper, they use IBD as another layer of filtering for data of two affected siblings from the same family under Mendelian recessive inheritance pattern. Right now, I have exome data from two families, in each of which there is ONLY one affected kid, and two healthy parents and another healthy sibling. The two families are from the same area, and two affected kids show the same symptons, thus we assume it's more likely to be recessive though we cannot rule out the possibility of de novo novel mutation. Under recessive model, what kinds of linkage/IBD analysis can I do to filter the variants from exome data? Are the two kids here supposed to inherit the same haplotype around disease-causing loci so that I can try to find out IBD region for them, though they're actually from two different families?

Actually I've tried to find intersection between variants of the two kids and interestingly found nothing. Then I decided to look at variants separately for each of them, then I may need further filtering.

Answer 1

[Zev.Kronenberg]

I would strongly suggest trying a VAAST search. One feature that VAAST has is the ability to filter SNPs/SNVs based on the model of inheritance. You can also pre-filter all of your variants to exclude sequencing error. The children should, for the most part, only have alleles seen in either parent.

Also pVAAST is coming shortly and that will allow you to directly supply the pedigree to gain extra power in your disease-gene hunt.