According to research published in Nature Communications by a KAUST-led team, scientists have developed an artificial intelligence tool named Unify to compare similar cell types across distant species and assess which animal study findings apply to human health. Cells from humans, mice, fish, flies, and worms perform similar biological functions even when their underlying genes diverge over millennia. Unify bridges this evolutionary gap by connecting 125 cell types across more than 700 million years of evolution, helping researchers identify which animal discoveries are most relevant to human health research.
Why Traditional Genetic Tools Struggle Across Evolution
Scientists frequently map biological activity using single-cell RNA sequencing to reveal active genes within individual cells, creating “cell trees” that chart cell types and their functions, according to KAUST findings. However, conventional comparison tools rely on direct one-to-one gene matches. These matches degrade as species diverge over time. Consequently, important biological similarities—such as connections between human and mouse immune cells or fish and fly neurons—risk being overlooked by standard genetic screening methods.
Overcoming Gene Divergence with Macrogenes
To fix these limitations, Unify evaluates the actual jobs that genes perform rather than hunting for literal gene matches. “Unify works with AI models that analyze protein sequences and scientific descriptions of gene functions,” explains Huawen Zhong, lead author and computational biologist at KAUST. Zhong notes that genes sharing specific functional traits are grouped into units called “macrogenes,” enabling the tool to spot cells performing similar jobs even when their individual genetic sequences no longer align.
Did you know? Unify functions much like a conceptual translator rather than a literal dictionary, seeking out shared biological meaning across species separated by over 700 million years of evolution, according to KAUST researchers.
Hidden Genetic Relationships Uncovered by AI
Unify successfully reconstructed relationships among 125 cell types spanning seven species, distinguishing between identical genes, similar genes that evolved new roles, and completely different genes that evolved independently to execute the same task. By analyzing cross-species immune cells, the AI identified shared defense tactics that eluded standard gene-comparison methods, according to KAUST data. In separate tests, Unify accurately predicted human blood cell responses to specific immune-signaling proteins using data drawn from mouse lymph-node immune cells, outperforming existing predictive methods across all genes.
Next Steps in Mapping Cell Evolution
The KAUST research team is currently expanding Unify to process additional layers of data, including gene regulation and exact cell positions within tissues. These upgrades will provide a comprehensive view of biological principles shared across life, according to university reports. Manuel Aranda, KAUST Professor of Marine Science, explains the broader stakes of the work: “A lot of what we know about human biology comes from studying animals like mice, but it is not always clear which findings carry over. Unify helps us identify which discoveries in model organisms are most likely to be relevant to humans, so research can be focused where it is most useful for understanding human health and disease.”
Frequently Asked Questions
What is the Unify AI tool?
Unify is a KAUST-developed AI tool published in Nature Communications that compares cell types across distant species by analyzing protein sequences and gene functions rather than relying solely on direct gene matches.
How many species and cell types does Unify connect?
According to KAUST researchers, Unify has connected 125 cell types across seven species separated by more than 700 million years of evolution.
Why is Unify useful for human health research?
By accounting for evolutionary changes in genes, Unify helps scientists determine which biological discoveries made in model organisms like mice are most relevant to human biology and disease.
Explore More: Want to stay updated on the latest breakthroughs in biotechnology and artificial intelligence? Subscribe to our newsletter or leave a comment below to join the discussion on the future of cross-species genetic research.
Keep reading