DIVERGE v4: A Platform for Large-Scale Analysis of Functional Divergence Across Multi-Gene Families

Gene duplication and subsequent divergence are fundamental drivers in the evolution of novel gene functions. The DIVERGE software, widely utilized for identifying amino acid residues critical to functional shifts between protein subfamilies, has been substantially upgraded to version 4 to address modern bioinformatics challenges. This major release features a modular Python package for enhanced workflow integration and a novel algorithm for scalable analysis of large gene families. Notably, we introduce DIVERGE-Chat, pioneering the application of a conversational artificial intelligence (AI) agent in the field of functional divergence, enabling intuitive, code-free exploration (https://pgx.zju.edu.cn/diverge). We have also developed a complementary interactive online web server that allows users to upload local files for interactive analysis, facilitating more flexible analysis. Furthermore, we curated phylogenetic data for 4,540 human gene families and conducted functional divergence analyses, generating a comprehensive resource accessible online. Collectively, these advancements significantly improve the accuracy, scalability, usability, and integration capabilities of functional divergence analysis, empowering broader research into protein evolution, functional specificity, and biomedical applications.