Conventional PCR-based sequencing of ancient or degraded DNA works by amplifying whatever DNA is present in a sample, using short primer sequences to initiate the amplification of target regions. The limitation is that PCR amplification is non-selective with respect to fragment length: it will amplify any template that includes the target sequences, whether that template is a short, highly fragmented piece of authentic ancient DNA or a long, intact piece of modern contaminating DNA. In a heavily contaminated sample, the contaminating DNA — being more intact — tends to dominate the PCR amplification, producing a result that reflects the contamination more than the authentic ancient material.
PEC MPS addresses this limitation in several ways. First, it uses a capture step that selects for the target sequences before sequencing, allowing more of the sequencing capacity to be devoted to the DNA of interest rather than to whatever other DNA happens to be present. Second, massively parallel sequencing produces very large numbers of individual sequence reads from each sample, allowing statistical analysis of the mixture of fragments present. Third, and most importantly for authenticating ancient DNA, it allows analysis of fragment length across the reads: authentic ancient DNA is expected to be highly fragmented, producing reads that are predominantly short, while modern contaminating DNA is expected to produce longer reads.
By examining the distribution of fragment lengths across the recovered sequences, the researchers could identify which sequences were consistent with authentic ancient material and which were more likely to be from recent contamination. This is a probabilistic analysis rather than a binary determination for each individual molecule, but across many thousands of sequence reads, it provides reliable statistical discrimination.