Researchers Develop hEVO Genomic Foundation Model
The new model interprets human genetic diversity by analyzing billions of haplotype sequences.
Updated on Oct. 21, 2026 in Life Sciences
Live Poll
Do you trust that new genomic AI models will improve the accuracy of medical disease diagnostics?
Scientists have introduced hEVO, a genomic foundation model designed to interpret complex human genetic diversity. The tool uses a massive training dataset to improve functional annotation and variant interpretation in human genomes.
Why it matters
The model enhances the ability to identify biological functions and pathogenic variants within human genomes. By leveraging shared sequence features and allelic differences, it provides a more efficient approach to understanding genetic variation.
The hEVO model was trained on 4.784 billion haplotype sequences derived from 63,809 individuals, selected from a total pool of 550,237 samples. It secured the top ranking in 18 of 19 primary performance benchmarks against other genomic models.
The details
By integrating shared sequence features and allelic differences without requiring clinical labels, hEVO identifies pathogenic classifications in ClinVar data. The model maintains high performance while utilizing fewer parameters than comparable billion-parameter genomic models.
Timeline
October 19, 2026: The paper describing the hEVO model was released.
The Big Picture
This development shifts the landscape of computational genomics by outperforming existing frameworks in variant interpretation. It builds on the utility of the ClinVar clinical variant database, providing a more robust tool for identifying pathogenic genetic classifications.
This genomic model could accelerate the discovery of disease-linked variants, potentially leading to faster and more accurate genetic testing in clinical settings. Researchers and diagnostic labs may benefit from improved efficiency in interpreting complex genomic data.
The takeaway
The hEVO model demonstrates how foundation models can be optimized to process massive biological datasets efficiently. This progress marks a significant step toward using machine learning to decipher the complex functional architecture of the human genome.
Further reading
Learn more about the latest innovations in Life Sciences.
More information
Access the complete research paper on hEVO model for technical specifications.
Live Poll
Do you trust that new genomic AI models will improve the accuracy of medical disease diagnostics?
