This is a very complex issue. Because we do not have large databases of human DNA, we are working off very small volunteer samples and extrapolating from them.
A good example would be the commercial genealogy DNA databases like 23 and Me. As more people provide samples the database is refined, so earlier reports on what your inheritance is likely to be are changed. They are working by comparing one set of samples with another from the same region, and the fewer people who are involved the less accurate their findings are likely to be.
We are just not at the stage where we can say there was definitely a wholesale massacre of one male population group in a specific area, or they died off, or went extinct for some reason. No scientist would say that. Popular writers might.