Genomes2Genes: A Progressive Cactus Gene Analysis Pipeline

Whole-genome alignments are fundamental resources in comparative genomics, because they place genes and their surrounding sequences in a shared evolutionary context. Progressive Cactus aligns large numbers of genomes without a reference genome and reconstructs ancestral sequences along the input phylogeny. However, its Hierarchical Alignment Format output has mainly been used for visualization, phylogenomics, and pangenome studies, and gene-level analyses of these alignments remain difficult. Here, we present Genomes2Genes, a pipeline that converts Progressive Cactus graph-based alignments into validated coding sequence alignments of extant species and their reconstructed ancestors. The pipeline extracts exon alignment blocks, handles atypical genomic jumps, stitches exons into transcripts, and reports start codons, stop codons, and frameshifting indels across species, with validation steps throughout and a base-by-base audit of the final alignments. It catches both biological and technical artifacts and can be run on any part of the genome. We anticipate that this pipeline will be useful in identifying and removing problematic regions used in the potential reconstruction of species or gene phylogenies. We developed and tested the pipeline on a whole-genome alignment of 21 Brassicaceae species and 2 outgroups and found that, among 45,491 lineage-specific genes, the pipeline supported 22,376 candidate de novo genes, of which 1,811 had complete evidence of enabler mutations. Genomes2Genes is publicly available and extends whole-genome alignments to gene-level questions in comparative genomics and evolutionary biology.