This suggests that post-mortem deaminations at cytosine residues contribute massively to the edit distance of alignments, limiting our ability to identify the fraction of endogenous reads that are most divergent.
Overall, we present a series of approaches that could substantially improve the ability to identify the true fraction of endogenous reads when shotgun sequencing aDNA with Illumina and Helicos next-generation sequencing platforms.
The fraction of endogenous reads is estimated by summing the number of reads identified in nucDNA and mtDNA and dividing by the total number of reads generated per channel (minus reads corresponding to spiking oligonucleotides in the specific case of spiking experiments).
In contrast, the fraction of new horse hits identified with this set of parameters consisted of reads of size longer than 34 nucleotides, suggesting that the option –n 0.03 could be used to improve our ability to identify endogenous reads without inflating false-positive rates (contrary to what has been observed for the most permissive edit distance).
This difference shows that oligonucleotide spiking reduced our capability to access endogenous ancient reads.
The fraction of endogenous Illumina reads identified is less sensitive to the mapping strategy.
Other sets of parameters explored have been found to improve our ability to identify endogenous sequence reads, however at extensive runtime costs, and therefore should be not be recommended on a systematic basis.
As endogenous aDNA reads have been shown to exhibit specific miscoding lesions that could be used as molecular signatures of post-mortem DNA damage, nucleotide misincorporation patterns [ 20] were used to assess the quality of the extra hits recovered with the different mapping strategies explored.
Using sequence similarity search algorithms it is thus possible to analyze DNA or RNA sequence data from a sample, compare the sequences against a set of reference sequences, and filter away all the endogenous ('expected') reads, leaving a small collection of sequences that do not appear to stem from the organism in question.
ubp8 was tagged after the stop codon of its endogenous open reading frame (ORF) with sequences encoding the Kanamycin resistance gene by a polymerase chain reaction (PCR -mediated strategy (Bahler et al. 1998) using a lithium acetate–based transformation PCR -mediatedeney and Boeke 1994).
