gnomAD has .vcf files publicly available that have variant data gathered from over 15k full genomes, for a university project I want to sample 1k of these genomes and get the variant data from those genomes. I am very new to bioinformatics in general and upon checking .vcf files and how they work, but my current conclusion is that this task is not achievable through filtering the vcf file in anyway since variants in these files have no direct connection to the genomes they occurred / were found in (and some fields of variants data hold information about the whole genome set like the percentage of occurrance in different geographical regions which makes removing some of the genomes from the data sets a bit nonesensical). But this task was given to me by a professor whom I believe is competent to realize the task was impossible if it is like I thought, so is there something I am missing? How can I go about filtering the 15k genome data of gnomad down to 1k?