That's because consumer DNA profiling outfits are trash, not because there's anything wrong with properly conducted DNA analysis. That's like saying the Large H

That's because consumer DNA profiling outfits are trash, not because there's anything wrong with properly conducted DNA analysis. That's like saying the Large Hadron Collider results must be wrong because the image on your old electron gun TV tube is blurry. The way out of this is to follow proper protocols for well conducted science. Peer review, multiple verification from different sites and by different labs. Unfortunately a bad result occasionally slips through, but it is possible to have high confidence in most well attested results. So the answer to how we know these things is that we have multiple samples including large sections of skulls and other major bones, from multiple sites, many of them analysed and verified by multiple different labs.