Tree-Conditioned Edit Flows for Ancestral Sequence Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sharafutdinov, Emil, André, Ingemar
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911649809563648
author Sharafutdinov, Emil
André, Ingemar
author_facet Sharafutdinov, Emil
André, Ingemar
contents Ancestral sequence reconstruction (ASR) aims to infer extinct protein sequences at internal nodes of a phylogenetic tree. Classical ASR methods are typically based on continuous-time Markov substitution models, but they treat sites largely independently and handle insertions and deletions only weakly or not at all. We introduce a tree-conditioned edit-flow model for variable-length ASR. Given two descendant sequences and their branch distances to a shared ancestor, the model reconstructs the ancestor through paired bidirectional edit trajectories constrained to agree on a common ancestral state. On a benchmark of experimentally evolved sequences with only context-independent substitutions, the model does not match the accuracy of the best classical method, yet still achieves reasonable performance despite being trained on natural sequences that include insertions, deletions, and substitutions. On a benchmark of natural homologous sequences with abundant insertions and deletions, the model most accurately localizes inferred evolutionary change.
format Preprint
id arxiv_https___arxiv_org_abs_2605_04119
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Tree-Conditioned Edit Flows for Ancestral Sequence Reconstruction
Sharafutdinov, Emil
André, Ingemar
Quantitative Methods
Machine Learning
Populations and Evolution
Ancestral sequence reconstruction (ASR) aims to infer extinct protein sequences at internal nodes of a phylogenetic tree. Classical ASR methods are typically based on continuous-time Markov substitution models, but they treat sites largely independently and handle insertions and deletions only weakly or not at all. We introduce a tree-conditioned edit-flow model for variable-length ASR. Given two descendant sequences and their branch distances to a shared ancestor, the model reconstructs the ancestor through paired bidirectional edit trajectories constrained to agree on a common ancestral state. On a benchmark of experimentally evolved sequences with only context-independent substitutions, the model does not match the accuracy of the best classical method, yet still achieves reasonable performance despite being trained on natural sequences that include insertions, deletions, and substitutions. On a benchmark of natural homologous sequences with abundant insertions and deletions, the model most accurately localizes inferred evolutionary change.
title Tree-Conditioned Edit Flows for Ancestral Sequence Reconstruction
topic Quantitative Methods
Machine Learning
Populations and Evolution
url https://arxiv.org/abs/2605.04119