For each reference, the directory provides a matrix of 1:1 or many:1 orthologs in tsv format (gzipped).
one2one_matrix.tsv.gz

The columns in this file correspond to the reference (column 1) and all query species (column 2-N).
Each row corresponds to a reference gene.
If a query species has a 1:1 or many:1 orthology relationship to the reference, the gene identifier is listed, otherwise '-'.

For 1:1 orthologs, TOGA2 assigns the same reference gene name to the query gene.
For many:1 orthologs, the query gene name is the comma-separated list of reference genes if only two or three reference genes correspond to query gene, otherwise the query gene name is one reference gene with a '_+' suffix. 

With the query gene names sequences, transcripts and other information can be extracted from the files in the TOGA2/$reference/vs_$query subdirectory.

====== REFERENCE =======
If you use the TOGA2 data, please cite:
Yury Malovichko et al. "Accurate, comprehensive gene annotation and ortholog identification across thousands of vertebrate genomes with TOGA2", https://www.biorxiv.org/content/10.64898/2026.06.30.735536v1