I am working through a SNP prioritisation pipeline designed by my lab to identify SNPs of functional potential for in vitro analysis. It involves the use of several tools (e.g., CADD, GWAVA, HaploReg, RegulomeDB, etc) as well as Ensembl and Variation Viewer as part of the annotation process.
Several of the SNPs I am manually annotating fall into fix patches which I understand take precedence over the primary assembly when performing analyses, however excluding Ensembl and Variation Viewer, the patches have not been integrated into other tools I am using to make predictions of functionality (e.g., cannot generate CADD scores with patch VCF, relies on primary assembly).
Given this, what is the best approach for performing the annotations? Do I ignore the patch and work with the primary assembly? Does the information from other tools referring to the primary assembly still stand? I don't want to pursue SNPs mapped to gene promoters in the primary assembly that have shifted in the patch, but at the same time I'm not sure whether the fact that the SNPs fall into patches makes the other functional information I have derived from other tools completely erroneous.
I am very new to bioinformatics, so any assistance would be greatly appreciated!