DeepVRegulome: DNABERT-based deep-learning framework for predicting the functional impact of short genomic variants on the human regulome

2025-11-12

AI Summary

Functional annotation of noncoding short variants and their clinical interpretation remain key bottlenecks in genomic medicine due to limited interpretability and accuracy. To address this, we propose a multi-model DNABERT ensemble framework: 700 DNABERT models are fine-tuned on large-scale ENCODE regulatory data, and integrated with variant effect scoring, motif perturbation analysis, attention visualization, and survival association testing—enabling precise, interpretable prioritization of splicing-regulatory and transcription factor binding site (TFBS) mutations. Applied to glioblastoma whole-genome sequencing data, our method identified 572 splice-disrupting variants and 9,837 TFBS-altering variants; among these, 1,352 were significantly associated with overall survival. Furthermore, we constructed a prognostic stratification model based solely on noncoding mutation features. This approach substantially improves both the accuracy and clinical translatability of noncoding variant functional interpretation.

Technology Category

• Machine Learning: Ensemble Methods
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Whole-genome sequencing (WGS) has revealed numerous non-coding short variants whose functional impacts remain poorly understood. Despite recent advances in deep-learning genomic approaches, accurately predicting and prioritizing clinically relevant mutations in gene regulatory regions remains a major challenge. Here we introduce Deep VRegulome, a deep-learning method for prediction and interpretation of functionally disruptive variants in the human regulome, which combines 700 DNABERT fine-tuned models, trained on vast amounts of ENCODE gene regulatory regions, with variant scoring, motif analysis, attention-based visualization, and survival analysis. We showcase its application on TCGA glioblastoma WGS dataset in prioritizing survival-associated mutations and regulatory regions. The analysis identified 572 splice-disrupting and 9,837 transcription-factor binding site altering mutations occurring in greater than 10% of glioblastoma samples. Survival analysis linked 1352 mutations and 563 disrupted regulatory regions to patient outcomes, enabling stratification via non-coding mutation signatures. All the code, fine-tuned models, and an interactive data portal are publicly available.

Problem

• Predicting functional impacts of non-coding genomic variants
• Prioritizing clinically relevant mutations in regulatory regions
• Identifying survival-associated mutations in glioblastoma datasets

Innovation

• DNABERT fine-tuned models predict variant impact
• Combines motif analysis with attention visualization
• Prioritizes survival-associated non-coding mutations