how to find if an observed missense SNV is true observation and significant in targeted sequencing of few (5 or less) genes.

Question

[oriolebaltimore]

Dear group, What are the methods/tests/algorithms that are available to find whether an observed missense/or any other SNV as significant variant among many cases (for a long genes, I am observing at least 10 missense and 10 synonymous variants).

If full genome or exome data is available, one would do mutsig2 etc. to test if observed SNV is by chance or ranks above background threshold. How do we deal with cases with VCF files and BAM files in hand. I am asking in the context of 4 genes, sequenced through amplicons from fluidigm/miseq setting.

Appreciate any input. thanks Adrian

[karl.stamm]

this is not clear: " find whether an observed missense/or any other SNV as significant variant among many cases " what are you trying to do ??

[WouterDeCoster]

Are you comparing patient samples with healthy control samples? Because that would make this an association analysis and you could either do a fisher exact test (e.g. in plink) or do a gene burden analysis/rare variant association analysis (e.g. in SKAT).