Search for: All records

Abstract Methods for rapidly inferring the evolutionary history of species or populations with genome-wide data are progressing, but computational constraints still limit our abilities in this area. We developed an alignment-free method to infer genome-wide phylogenies and implemented it in the Python package TopicContml. The method uses probabilistic topic modeling (specifically, Latent Dirichlet Allocation) to extract topic frequencies from k-mers, which are derived from multilocus DNA sequences. These extracted frequencies then serve as an input for the program Contml in the PHYLIP package, which is used to generate a species tree. We evaluated the performance of TopicContml on simulated datasets with gaps and three biological datasets: 1) 14 DNA sequence loci from two Australian bird species distributed across nine populations, 2) 5162 loci from 80 mammal species, and 3) raw, unaligned, nonorthologous PacBio sequences from 12 bird species. We also assessed the uncertainty of the estimated relationships among clades using a bootstrap procedure. Our empirical results and simulated data suggest that our method is efficient and statistically robust. 

Phylogenetic trees are fundamental for understanding evolutionary history. However, finding maximum likelihood trees is challenging due to the complexity of the likelihood landscape and the size of tree space. Based on the Billera-Holmes-Vogtmann (BHV) distance between trees, we describe a method to generate intermediate trees on the shortest path between two trees, called pathtrees. These pathtrees give a structured way to generate and visualize part of treespace. They allow investigating intermediate regions between trees of interest, exploring locally optimal trees in topological clusters of treespace, and potentially finding trees of high likelihood unexplored by tree search algorithms. We compared our approach against other tree search tools (P aup *, RA x ML, and R ev B ayes ) using the highest likelihood trees and number of new topologies found, and validated the accuracy of the generated treespace. We assess our method using two datasets. The first consists of 23 primate species (CytB, 1141 bp), leading to well-resolved relationships. The second is a dataset of 182 milksnakes (CytB, 1117 bp), containing many similar sequences and complex relationships among individuals. Our method visualizes the treespace using log likelihood as a fitness function. It finds similarly optimal trees as heuristic methods and presents the likelihood landscape at different scales. It found relevant trees that were not found with MCMC methods. The validation measures indicated that our method performed well mapping treespace into lower dimensions. Our method complements heuristic search analyses, and the visualization allows the inspection of likelihood terraces and exploration of treespace areas not visited by heuristic searches.