Published in 2026, Species-OT is a mathematical method developed by researchers at Kyoto University that uses optimal transport to compare gene expression patterns and map gene correspondences across different species, bridging animal model data and human biology.
For instance, genes crucial for forming germ cells—which eventually develop into eggs and sperm—differ between mice and humans. Traditionally, scientists compare DNA sequences to establish evolutionary relationships, but that approach struggles to answer a vital functional question: which genes perform similar regulatory work within specific cell types across different organisms?
Kyoto University Researchers Calculate Gene Expression Geometry
A team at Kyoto University’s Institute for the Advanced Study of Human Biology (WPI-ASHBi) analyzed how genes are expressed in relation to one another.
The team constructed Species-OT, a data-driven method designed to compare patterns of gene expression to identify genes that may play similar roles across species. The research centered its analysis on genes that code for transcription factors, anticipated to constitute the highly conserved backbone of gene regulatory networks. By treating gene expression vectors as points, the algorithm places genes with similar expression patterns closer together. This forms a geometric shape characterizing the dataset. Using Gromov-Wasserstein optimal transport, the mathematical framework compares how these points are arranged relative to one another, identifying candidate gene correspondences and measuring how much the overall gene expression patterns differ between datasets.
Primordial Germ Cell Induction
To test the method, the team examined gene expression during the induction of primordial germ cell-like cells (PGCLCs) derived from mouse and human pluripotent stem cells. Among the top predictions generated by Species-OT were pairs of mouse and human genes that previous studies had shown to play similar roles.
Translating Animal Models
Lead author Dr. Yuya Tokuta explained that comparing the mathematical structures underlying genomic, epigenomic, and gene expression data across species could help understand where and how humans differ from other organisms, as well as what they have in common. Eventually, Species-OT has the potential to bridge animal research and human pathology by highlighting genes that might function similarly across different organisms.
While the evaluation of gene correspondences in this study focused mainly on a selected set of genes involved in germ cell specification, the method establishes a foundation for comparative genomics that goes beyond conventional phylogenetic trees.