I am working with WES data for detection of somatic variants and I have used two variant callers because no variant caller is complete in itself. I have used GATK Haplotypecaller for small variants like snp,ins,del etc. And for large variants I have used pindel (specially for SVs detection). Now I have two vcf file for each patients, one from gatk and another from pindel. Although I know various vcf-merging tools but is it right to merge two different vcf file from different callers. Do they give right and proper merged vcf file. Are those vcf-merging tools SV-aware and consider only trivial call for the same variant represented differently in different callers. I know these are too many questions for a single thread but all these are related to each other. I am in dilemma whether I should merge them or use them separately.
The merge utility from SURVIVOR seems to be what you are looking for. I did not try it yet, but it seems to be especially designed to handle SV.
The simplest thing to do would probably to first merge your patients files together for each tool and have one multi-patient file per variant caller. Then I would merge the multi-sample files of the different tools.
Here is a detailed blog post describing the different challenges associated with merging structural variants from different callers and explaining the steps they took to address it. All the code to reproduce the pipeline is available on a github repo.