Hyperbolic Genome Embeddings

2025-07-29

AI Summary

Existing genomic modeling approaches struggle to align machine learning inductive biases with the hierarchical, tree-like structure inherent in biological evolution. Method: We propose HyperGen, the first hyperbolic genome embedding framework that introduces hyperbolic convolutional neural networks for DNA sequence modeling—implicitly encoding evolutionary relationships without requiring explicit phylogenetic trees. The framework integrates hyperbolic exponential/logarithmic mappings, high-dimensional hyperbolic embeddings, and interpretability analysis techniques, and introduces a novel benchmark dataset incorporating transposable elements. Contribution/Results: HyperGen outperforms Euclidean baselines on 37 of 42 downstream tasks; it surpasses state-of-the-art DNA language models on 7 GUE (Genomic Understanding Evaluation) tasks while using 1–2 orders of magnitude fewer parameters and no pretraining. We further introduce the first quantitative metric for hyperbolicity assessment in genomics, empirically validating the intrinsic suitability of hyperbolic geometry for modeling genomic function and regulation.

Technology Category

Machine Learning: Evolutionary LearningSearch and Optimization: Evolutionary ComputationNatural Language Processing: Generation

Application Category

Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphsSemantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsWeb Mining and Content Analysis: Models for Web evolution

Abstract

Current approaches to genomic sequence modeling often struggle to align the inductive biases of machine learning models with the evolutionarily-informed structure of biological systems. To this end, we formulate a novel application of hyperbolic CNNs that exploits this structure, enabling more expressive DNA sequence representations. Our strategy circumvents the need for explicit phylogenetic mapping while discerning key properties of sequences pertaining to core functional and regulatory behavior. Across 37 out of 42 genome interpretation benchmark datasets, our hyperbolic models outperform their Euclidean equivalents. Notably, our approach even surpasses state-of-the-art performance on seven GUE benchmark datasets, consistently outperforming many DNA language models while using orders of magnitude fewer parameters and avoiding pretraining. Our results include a novel set of benchmark datasets--the Transposable Elements Benchmark--which explores a major but understudied component of the genome with deep evolutionary significance. We further motivate our work by exploring how our hyperbolic models recognize genomic signal under various data-generating conditions and by constructing an empirical method for interpreting the hyperbolicity of dataset embeddings. Throughout these assessments, we find persistent evidence highlighting the potential of our hyperbolic framework as a robust paradigm for genome representation learning. Our code and benchmark datasets are available at https://github.com/rrkhan/HGE.

Problem

• Aligning machine learning biases with biological evolutionary structure
• Improving DNA sequence representations using hyperbolic CNNs
• Outperforming Euclidean models in genome interpretation benchmarks

Innovation

• Hyperbolic CNNs for DNA sequences
• No explicit phylogenetic mapping needed
• Outperforms Euclidean models and DNA language models