Genome-wide genealogies for thousands of samples

Genealogies describe how DNA samples are related through most recent common ancestors back in time. In principle, they are the best attainable record of the genetic past of a sample of DNA sequences and therefore, if known, simplify and substantially enhance any inferences we make about our shared genetic past.

RELATE makes inference of genealogical histories for large sample sizes possible, a problem that despite its importance had previously remained infeasible for all but the smallest datasets. RELATE scales to >10,000 sequences and improves on accuracy in scenarios with realistic levels of errors.

Tools for downstream analyses

RELATE comes with tools for downstream analyses of inferred genealogies to tackle a broad range of applications, including for estimating demographic history, mutation rates, and detection of positively selected mutations. Genealogy-based inferences can be more powerful or more accurate than alternative specialist methods.

Main benefits include:

• Scalable to thousands of whole-genome sequences
• Very powerful inference framework applicable to questions related to evolutionary biology, population genetics, and genetic disease
• All inferences are derived from the same genealogies leading to better consistency across different applications
• Improved accuracy in areas of realistic errors

Commercialisation

Oxford University Innovation is actively seeking partners that may wish to licence this software. For further details, please see the research groups webpage https://myersgroup.github.io/relate/