Willerslev’s team spent approximately fourteen years — from the initial access to the hair around 2007 until the publication of the study in 2021 — developing and refining a method that could work with the material they had.
—
## Autosomal DNA: A Different Kind of Inheritance
The solution the team developed relied on autosomal DNA — the genetic material that makes up the vast majority of the human genome, inherited from both parents rather than exclusively from one parent’s line.
Autosomal DNA is what people typically think of when they think of genetic inheritance: the mix of material from mother and father, and from all four grandparents, and from all eight great-grandparents, and so on backward through generations. Each generation, roughly half of the autosomal DNA from each parent is passed on. By the generation of a great-grandchild, only about one-eighth to one-sixteenth of any particular great-grandparent’s autosomal DNA is expected to appear, though the exact proportion varies because of the randomness of how DNA segments are shuffled in reproduction.
As Willerslev explained in public statements about the study, “Autosomal DNA is our non-gender-specific DNA” — meaning it can identify relationships through either the paternal or maternal side, making it potentially applicable in cases where the lineage runs through an unconventional combination of parent-lines.
The challenge with degraded ancient DNA is that the fragments are short and sparse. When you are trying to identify the degree of relationship between two people using autosomal DNA, you need to find segments of DNA shared between them that are long enough to indicate shared descent rather than random coincidence. Short fragments, or low-coverage data, make this harder. The mathematics of relatedness detection at three generations of separation — the great-grandparent relationship — are challenging even with modern, high-quality samples. With a degraded century-old hair sample, they are considerably more demanding.