The Indus Valley Civilization is as mysterious as Atlantis, except that we know the Indus Valley Civilization was very real. More than 5,000 years ago, the people of the Indus Valley had planned cities with running water — even household toilets, and roads that would put the Romans to shame, more than 2000 years before Rome was founded. But beyond the remains of their public works, we know almost nothing about who these people were — until now. The largest ever study of human DNA, covering thousands of years and thousands of miles, tells us exactly who they were, revealing the actual genome of many ancient people. The findings appear in a pair of papers in the current issues of Science and Cell. One of the study's authors, Professor David Reich of Harvard Medical School, spoke with WGBH Radio's Arun Rath. This transcript has been edited for clarity.

Arun Rath: So before we get into the details of this, I have to ask, I guess, a technical question: How do you get the DNA from humans going back thousands of years?

David Reich: Well it's something that's only become possible in the last 10 years. It starts with a bit of bone or a tooth, ground up. We extract and release the DNA in a solution that removes proteins and minerals, and we convert it into a form that can be sequenced. And using one of these sequencers that have dropped the cost of sequencing by literally a million fold in the last two decades, we're able to obtain a little bit of DNA from these individuals.

Rath: So for the Indus Valley people, there's some historical baggage on this already. You know, I'm a Hindu, and I knew growing up of this idea that ancient Indians actually came from somewhere else, supposedly from from Europe. What does the actual DNA tell us about who these people were?

Reich: So we have 12 individuals who are associated with the Indus Valley Civilization in our study from about 524 individuals for whom we obtained data. So one of those individuals is buried actually in an Indus Valley site, and 11 of those individuals are migrants in other sites in Iran and to the north, in Turkmenistan.

And what we can see in these individuals is they had no ancestry at all, as far as we can tell, from the steppe north of the Black and Caspian seas. So the ancestry of these Indus Valley Civilization people, as far as we can tell, is entirely local to South Asia for thousands of years prior to the individuals we sequenced.

Rath: There is, unfortunately, political baggage when you're doing analysis of this sort. The Indus Valley civilization is in present-day Pakistan, which obviously doesn't have the best relations with present-day India. There are racial theories even going back to the Nazis, who talked about the original Aryan people that were pure before they migrated and went to other places. Is this something that, when you do work like this, do those external political factors bear on you at all?

Reich: So I think that the effect, in my view, of ancient DNA studies has always been to overthrow people's preconceived notions of what occurred. The spread of these people is almost certainly responsible for the spread of the languages spoken both in almost every European country, Indo-European languages, and also in India and Iran and Armenia. So it's really been a mystery for more than 200 years why these languages across this vast region are so similar to each other and what movements of people could have spread them. We now know the answer to that, but it's not what people who have been using these ideas in archaeology for political ends would have thought. For example, the Germans thought that Indo-European languages were spread from the area of Germany and Poland.

Rath: You have this almost head-spinning amount of data about ancient humans. Are there any big areas of mystery left? Is there something for you to go after next?

Reich: Well this is just the tip of the iceberg in terms of the ancient DNA revolution. One thing that we know now, just in the last few years, that we didn't know before is how common profound mixture is in human population history. The Indus Valley Civilization, people that we have data from, are mixed of two earlier sources. They then mixed after 4,000 years ago, we now know, with groups to the north from the steppe and with groups to the southeast who are related to southeast Asian hunter-gatherers, producing two later populations. And it's a mixture of those two mixed populations that form South Asians today. That's not unique. You see that pattern in Europe, you see that pattern in the Near East, you see that pattern in East Asia, you see that pattern in Africa. Mixture is a common theme of our history. It's a common feature of who we are. Nobody is pure, and that's a new lesson that's really emerged from this study.