PhyloGFN: Phylogenetic inference with generative flow networks
This work addresses the problem of inferring evolutionary relationships in computational biology, offering an incremental improvement by applying a known method (GFlowNets) to a specific domain bottleneck.
The paper tackles the challenge of phylogenetic inference from sequence data by using generative flow networks (GFlowNets) to sample from multimodal posterior distributions over tree topologies and evolutionary distances, achieving competitive marginal likelihood estimation and a closer fit to target distributions than state-of-the-art variational inference methods on real benchmark datasets.
Phylogenetics is a branch of computational biology that studies the evolutionary relationships among biological entities. Its long history and numerous applications notwithstanding, inference of phylogenetic trees from sequence data remains challenging: the high complexity of tree space poses a significant obstacle for the current combinatorial and probabilistic techniques. In this paper, we adopt the framework of generative flow networks (GFlowNets) to tackle two core problems in phylogenetics: parsimony-based and Bayesian phylogenetic inference. Because GFlowNets are well-suited for sampling complex combinatorial structures, they are a natural choice for exploring and sampling from the multimodal posterior distribution over tree topologies and evolutionary distances. We demonstrate that our amortized posterior sampler, PhyloGFN, produces diverse and high-quality evolutionary hypotheses on real benchmark datasets. PhyloGFN is competitive with prior works in marginal likelihood estimation and achieves a closer fit to the target distribution than state-of-the-art variational inference methods. Our code is available at https://github.com/zmy1116/phylogfn.