Maximizing the acquisition of unique reads in non-invasive capture sequencing experiments

Non-invasive samples as a source of DNA are gaining interest in genomic studies of endangered species. However, their complex nature and low endogenous DNA content hamper the recovery of good quality data. Target capture has become a productive method to enrich the endogenous fraction of non-invasive samples, such as feces, but its sensitivity has not yet been extensively studied. Coping with fecal samples with an endogenous DNA content below 1% is a common problem when prior selection of samples from a large collection is not possible. However, samples classified as unfavorable for target capture sequencing might be the only representatives of unique specific geographical locations or to answer the question of interest. To explore how library complexity may be increased without repeating DNA extractions and generating new libraries, here we have captured the exome of 60 chimpanzees (Pan troglodytes) using fecal samples with very low proportions of endogenous content (< 1%). Our results indicate that by performing additional hybridizations of the same libraries, the molecular complexity can be maintained to achieve higher coverage. Also, whenever possible, the starting DNA material for capture should be increased. Lastly, we have specifically calculated the sequencing effort needed to avoid exhausting the library complexity of enriched fecal samples with low endogenous DNA content. This study provides guidelines, schemes and tools for laboratories facing the challenges of working with non-invasive samples containing extremely low amounts of endogenous DNA.