Multi-scale haplotype signatures distinguish among evolutionary processes
Natural selection embeds distinct genomic signatures as populations adapt to environmental pressures. Machine learning approaches, particularly deep neural networks, have emerged as powerful tools to detect adaptive signatures from genomic data. While machine learning has dramatically improved our ability to extract these signals from sequence variation, distinguishing among multiple evolutionary…
Natural selection leaves unique genomic imprints as populations adjust to environmental stresses. Recent advances in machine learning, especially deep neural networks, have proven adept at identifying adaptive signals hidden within genomic data. Yet, classifying these signals to discern between different evolutionary mechanisms proves difficult.
Researchers have now developed SHADES, a deep learning model that leverages data on sequence differences and their frequencies across various spatial scales to discern evolutionary signatures at multiple genomic levels. Experiments demonstrated SHADES’ ability to differentiate between neutral evolution, selective sweeps, balancing selection, and adaptive introgression, and it outperformed existing methods in both classification accuracy and power.
Furthermore, SHADES maintained strong performance even when faced with various confounding factors, showing only minor drops in accuracy. When applied to real-world haplotypic data from European humans, SHADES confirmed known patterns of selective sweeps, balancing selection, and adaptive introgression, and also uncovered new genes linked to immune response, inflammation regulation, and epithelial barrier function.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
