The above figure is from Population and genomic lessons from genetic analysis of two Indian populations. What you see here is that two Indian Hindu populations from the north and south of the subcontinent have clearly elevated stretches of genomic homozygosity in comparison to the classic Northwest European population of whites from Utah. This is interesting because the social practices of the two groups here are quite different. Some South Indian Hindus practice consanguineous marriage; e.g., first cousins or uncle-niece. This is evident in some individuals in the data set. But North Indian Hindus traditionally enforce significant exogamy among relations via the gotra system and seeking partners outside natal villages. And yet the genomic evidence indicates a relatively small effective population. That’s because though North Indian Hindus practice exogamy on the scale of families, they nevertheless usually marry within a local caste. The effect of this genomically was one of the less trumpeted findings of the 2009 paper Reconstructing Indian Population History. India may have a very large population, but the genealogical history of many of its people is sharply delimited. This recent paper uses exomes, and I think clinches the finding.
Second, two data sets that I stumbled upon in case you don’t know which are in VCF and phased Beagle format (though the newest release of Beagle uses VCF anyhow):
Singapore Sequencing Malay, 100 Malays.
Singapore Sequencing Indian. 36 individuals. Mostly South Indian Tamil.