Estimating genome-wide phylogenies using probabilistic topic modeling

Inferring the evolutionary history of species or populations with genome-wide data is gaining ground, but computational constraints still limit our abilities in this area. We developed an alignment-free method to infer the genome-wide species tree and implemented it in the Python package TopicContml. The method uses probabilistic topic modeling (specifically, Latent Dirichlet Allocation or LDA) 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 our method with biological and simulated data sets: a data set with 14 DNA sequence loci from 78-92 haplotypes from two Australian bird species distributed in 9 populations; a second data set of 5162 loci from 80 mammal species; and a third data set of 67317 autosomal loci and 4157 X-chromosome loci of 6 species in the Anopheles gambiae ..., PacBio HiFi reads were generated from DNA isolated from four birds: a Woodhouse's Scrub-Jay (Aphelocoma woodhouseii, specimen MCZ:Orn:365326); a Yucatan Jay (Cyanocorax yucatanicus, specimen MCZ:Orn:365269); Tataupa Tinamou (Crypturellus tataupa); and White-throated Tinamou (Tinamus guttatus). DNA was isolated using Qiagen Magattract, and the concentration was estimated using a TapeStation. The DNA of the two tinamous was sequenced on a PacBio HiFi Revio DNA sequencer at the Faculty of Arts & Sciences Bauer Core, Harvard University, and the DNA from the two jays was sequenced on a PacBio HiFi machine at the DNA Sequencing & Genotyping Center at the University of Delaware., , # Data from: Estimating genome-wide phylogenies using probabilistic topic modeling

[https://doi.org/10.5061/dryad.73n5tb36r](https://doi.org/10.5061/dryad.73n5tb36r)

## Description of the data and file structure

The data consists of two fasta.gz files, which are files containing sequence data from each species. These files were converted from fastq.gz files, which is the form in which the original data was produced after post-processing from the PacBio HiFi sequencer. Fasta.gz is a standard format for DNA sequence data produced from genomes.

### Files and variables

#### File: two\_birds\_Pacbio.tar.gz

**Description:** The file consists of two tarred and gzipped files, as follows:

* Ctat_100k_seqtk_Pacbio_reads.fasta.gz -100,000 PacBio HiFi reads from Crypturellus tataupa

- Tgut_100k_seqtk_Pacbio_reads.fasta.gz - 100,000 PacBio HiFi reads from Tinamus guttatus

The file was compiled with this command line:

tar cvf two_birds_Pacbio.tar.gz Ctat_100k_seqtk_Pacbio_reads.fasta.gz Tgu...