If we let ancient DNA dictate the terms in the Proto-Indo-European (PIE) homeland debate ahead of historical linguistics and archeology, then, as far as I can see, there are two or three realistic options for the location of the said homeland. Here they are, in order of my own preference:
1) The Don-Caspian steppe around 4,300 BCE (see here). The ancestors of the Hittites and other Anatolian speakers also came from this homeland and entered Anatolia via the Balkans (or, less likely, the Caucasus) in fairly small groups sometime between 4,000 and 2,000 BCE. A lot of samples from Bronze Age Anatolia are needed to confirm or debunk the presence of steppe ancestry there.
2) The eastern Balkans during the peak of the ostentatious Copper Age in the region. Proto-Indo-European developed in the wealthy Chalcolithic communities of the western Black Sea coast and quickly spread both into the steppes and Anatolia via elite and trade contacts, and thus with minimal gene flow. Proto-Indo-European minus Anatolian, or PNIE, then spread from Eastern Europe during the Bronze Age with the mass migrations of the Yamnaya and closely related populations. A lot of samples from Chalcolithic western Anatolia are needed to confirm or debunk that people moved from the Balkans into Anatolia at this time.
3) Transcaucasia and/or nearby around 10,000 BCE. Proto-Indo-European, or rather Indo-Hittite, is much older than generally accepted, and came from the Epipaleolithic northern Near East. It was introduced into the steppes by foragers of the so called Caucasus Hunter-Gatherer (CHG) type, where it eventually became Proto-Indo-European minus Anatolian, or PNIE. Proto-Anatolian was spoken by closely related CHG-like foragers who stayed in the northern Near East.
Admittedly, that last theory is way out there, and at the moment, has about as much chance of being accepted by most historical linguists as Out-of-India. But the one advantage that is has over the other two proposals is that it doesn’t need any additional sampling of ancient DNA.
I’ll probably get grilled in the comments why I didn’t include a proposal with the Maykop culture as the PIE community, or at least the Indo-Europeanizing agent in the steppe. Honestly, after seeing the ancient DNA from a wide range of Maykop remains courtesy of Wang et al., I think the chances that Maykop was an Indo-European-speaking culture are low. Indeed, both the Maykop genome-wide data and uniparental markers scream «Northwest Caucasian» to me.
Also, if the Caucasus was the PIE homeland, or even a major expansion point for early Indo-European languages, then considering its widely accepted status as a linguistic hotspot and refuge, it’s fair to expect that it should still harbor at least one highly diverged Indo-European language. Is there any evidence that it ever did?
Below is an interactive poll. Please vote for one of the three options and feel free to let us know in the comments why you made the choice that you did. I might add more options to the poll if compelling reasons are given in the comments to do so.