Hello,
I am new in ngs data alaysis. and I am not sure how to continue. I am trying to look at GATK best practices to do my analysis in the right way. I got fastq data. With all steps that follows I came to VQSR. After this step it says (in GATK best practices): This process is the last step of the pipeline and the final VCF file can be used for other genomic analyses. So,my question is how to use these vcf files to get variants that are connected with a disease. I am thinking about using Galaxy, Exomiser or SnpEff. And the next question..I will probably get some variants that are not reported very often, so I do not know where to check my results after that. For exmple: I will get a SNP that is pathogenic, but how could I know that this is really pathogenic variant. I supose I should not blindly rely on a tool? I have to prove this variants but how?
Thank you very much for your help
Eventually all computational predictions are only approximations or suggestions. You will have to perform functional experiments (e.g. a knockout, CRISPR-like experiment, overexpression of the gene with that variant) or see whether you can show a credible correlation between specimen that carry this variant (e.g. a cohort of patients) and their phenotype (e.g. presence of a disease). Alternatively, see whether there is published work on that variant.
It's nice to read this topic because that was my question a few years ago! So, after running GATK, you can upload your vcf file into Ensembl Vep and check the results. It will give you a complete annotation of your variants. Check public databases like GnomAD and predictors like Sift and Polyphen. In the beginning, this will look a little bit boring job but is the most important part of your job!
Thank you very much. I will do as you sugested.