Search papers, labs, and topics across Lattice.
The authors introduce ESMDisPred, a novel deep learning architecture for predicting intrinsically disordered proteins (IDPs) that leverages ESM2 embeddings and structural information from the PDB. By integrating sequence embeddings with structural features and employing feature engineering techniques like terminal residue encoding, the model achieves state-of-the-art performance on CAID3 benchmarks. The core of ESMDisPred is a hybrid CNN-Transformer architecture designed to capture both local sequence motifs and long-range dependencies.
By fusing protein language model embeddings with structural data, ESMDisPred achieves state-of-the-art accuracy in predicting intrinsically disordered proteins, a feat that could accelerate drug discovery and structural biology.
Intrinsically disordered proteins (IDPs) lack stable three-dimensional structures, yet play vital roles in key biological processes, including signaling, transcription regulation, and molecular scaffolding. Their structural flexibility presents significant challenges for experimental characterization and contributes to diseases such as cancer and neurodegenerative disorders. Accurate computational prediction of IDPs is important for advancing research and drug discovery, structural biology, and protein engineering. In this study, we introduce ESMDisPred, a novel structure-aware disorder predictor that builds on the representational power of Evolutionary Scale Modeling-2 (ESM2) protein language models. ESMDisPred integrates sequence embeddings with structural information from the Protein Data Bank (PDB) to deliver state-of-the-art prediction accuracy. Model performance is further enhanced through feature engineering strategies, including terminal residue encoding, statistical summarization, and sliding-window analysis. To capture both local sequence motifs and long-range dependencies, we designed a hybrid CNN-Transformer architecture that balances convolutional efficiency with the representational power of self-attention. On CAID3 benchmarks, our latest model achieves ROC-AUC 0.895, AP 0.778, and a max F1 of 0.759, outperforming recent methods. Our results highlight the importance of integrating protein language model embeddings with explicit structural information for improved disorder prediction.
