Likelihoods for a general class of ARGs under the SMC DOI Creative Commons
Gertjan Bisschop, Jerome Kelleher, Peter L. Ralph

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 27, 2025

Ancestral recombination graphs (ARGs) are the focus of much ongoing research interest. Recent progress in inference has made ARG-based approaches feasible across range applications, and many new methods using inferred ARGs as input have appeared. This on long-standing problem ARG proceeded two distinct directions. First, Bayesian under Sequentially Markov Coalescent (SMC), is now practical for tens-to-hundreds samples. Second, approximate models heuristics can scale to sample sizes three orders magnitude larger. Although these heuristic reasonably accurate metrics, one significant drawback that they estimate do not topological properties required compute a likelihood such SMC present-day formulations. In particular, typically precise details about events, which currently likelihood. this paper we present backwards-time formulation derive straightforward definition general class model. We show does require events be estimated, robust presence polytomies. discuss possibilities opens.

Language: Английский

Likelihoods for a general class of ARGs under the SMC DOI Creative Commons
Gertjan Bisschop, Jerome Kelleher, Peter L. Ralph

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 27, 2025

Ancestral recombination graphs (ARGs) are the focus of much ongoing research interest. Recent progress in inference has made ARG-based approaches feasible across range applications, and many new methods using inferred ARGs as input have appeared. This on long-standing problem ARG proceeded two distinct directions. First, Bayesian under Sequentially Markov Coalescent (SMC), is now practical for tens-to-hundreds samples. Second, approximate models heuristics can scale to sample sizes three orders magnitude larger. Although these heuristic reasonably accurate metrics, one significant drawback that they estimate do not topological properties required compute a likelihood such SMC present-day formulations. In particular, typically precise details about events, which currently likelihood. this paper we present backwards-time formulation derive straightforward definition general class model. We show does require events be estimated, robust presence polytomies. discuss possibilities opens.

Language: Английский

Citations

0