I'm currently attempting to lift over variants from a .vcf file mapped to the hg18 assembly to the chimpanzee panTro4 assembly (UCSC nomenclature used here). I'm using the picardtools LiftoverVcf to accomplish this, as well as the .chain file available on UCSC.
I'm noticing that only about 10% of variants are being lifted over, the rest are ending up in the 'regected' variants file. I'm wondering if this should be expected when lifting over from one organism to another, or if this is a strange occurrence.
Note: the original .vcf (which used hg18 assembly as the reference) was produced with short reads from chimpanzees. So, my guess (if it this is not a bug in the picardtools program) is that the vast majority of "variants" picked up in the .vcf are, in fact, just fixed differences between human and chimpanzee.
I'm noticing that only about 10% of variants are being lifted over, the rest are ending up in the 'regected' variants file. I'm wondering if this should be expected when lifting over from one organism to another, or if this is a strange occurrence.
Note: the original .vcf (which used hg18 assembly as the reference) was produced with short reads from chimpanzees. So, my guess (if it this is not a bug in the picardtools program) is that the vast majority of "variants" picked up in the .vcf are, in fact, just fixed differences between human and chimpanzee.