Computational filtering can remove reads that do not match the expected short-fragment profile of ancient DNA.

The characteristic chemical damage pattern of ancient DNA — specifically, the increased frequency of cytosine-to-thymine mismatches at the ends of DNA fragments, resulting from cytosine deamination over time — is another authentication criterion.

Genuine ancient sequences show this damage pattern; modern contamination does not.

Computational analysis of the sequencing data can identify and authenticate sequences on this basis.

After authentication, the surviving ancient reads are mapped to a reference human genome.

Each sequenced fragment is aligned to the position in the reference genome that it most closely matches.