I have sequencing tumour data of human species. i dont have normal samples only tumour samples. I followed all the steps for SNP calling such as Mapping of all samples , added read group detail, merge and mark duplicates accordingly and done GATK snp analysis. I would like to predict the Novel SNPs by doing comaprision with already SNPs available under TCGA and COSMIC database. How to do comparision? Is it possible to do comaprision at SNP calling step using GATK ?