11 January 2024 in Biology Methods and Protocols
Proteins are complex biomolecules essential for numerous biological processes, making them crucial targets for advancements in molecular biology, medical research, and drug design. Understanding their intricate, hierarchical structures, and functions is vital for progress in these fields. To capture this complexity, we introduce Multimodal Protein Representation Learning (MPRL), a novel framework for symmetry-preserving multimodal pretraining that learns unified, unsupervised protein representations by integrating primary and tertiary structures. MPRL employs Evolutionary Scale Modeling (ESM-2) for sequence analysis, Variational Graph Auto-Encoders (VGAE) for residue-level graphs, and PointNet Autoencoder (PAE) for 3D point clouds of atoms, each designed to capture the spatial and evolutionary intricacies of proteins while preserving critical symmetries. By leveraging Auto-Fusion to synthesize joint representations from these pretrained models, MPRL ensures robust and comprehensive protein representations. Our extensive evaluation demonstrates that MPRL significantly enhances performance in various tasks such as protein–ligand binding affinity prediction, protein fold classification, enzyme activity identification, and mutation stability prediction. This framework advances the understanding of protein dynamics and facilitates future research in the field. Our source code is publicly available at https://github.com/HySonLab/Protein_Pretrain.
19 July 2023 in The Journal of Chemical Physics
Contemporary graph learning algorithms are not well-suited for large molecules since they do not consider the hierarchical interactions among the atoms, which are essential to determining the molecular properties of macromolecules. In this work, we propose Multiresolution Graph Transformers (MGT), the first graph transformer architecture that can learn to represent large molecules at multiple scales. MGT can learn to produce representations for the atoms and group them into meaningful functional groups or repeating units. We also introduce Wavelet Positional Encoding (WavePE), a new positional encoding method that can guarantee localization in both spectral and spatial domains. Our proposed model achieves competitive results on three macromolecule datasets consisting of polymers, peptides, and protein-ligand complexes, along with one drug-like molecule dataset. Significantly, our model outperforms other state-of-the-art methods and achieves chemical accuracy in estimating molecular properties (e.g., highest occupied molecular orbital, lowest unoccupied molecular orbital, and their gap) calculated by Density Functional Theory in the polymers dataset. Furthermore, the visualizations, including clustering results on macromolecules and low-dimensional spaces of their representations, demonstrate the capability of our methodology in learning to represent long-range and hierarchical structures. Our PyTorch implementation is publicly available at https://github.com/HySonLab/Multires-Graph-Transformer.
7 August 2024 in IEEE Transactions on Evolutionary Computation
Directed evolution, a strategy for protein engineering, optimizes protein properties (i.e., fitness) by a rigorous and resource-intensive process of screening or selecting among a vast range of mutations. By conducting an in silico screening of sequence properties, machine learning-guided directed evolution (MLDE) can expedite the optimization process and alleviate the experimental workload. In this work, we propose a general MLDE framework in which we apply recent advancements of Deep Learning in protein representation learning and protein property prediction to accelerate the searching and optimization processes. In particular, we introduce an optimization pipeline that utilizes Large Language Models (LLMs) to pinpoint the mutation hotspots in the sequence and then suggest replacements to improve the overall fitness. Our experiments have shown the superior efficiency and efficacy of our proposed framework in the conditional protein generation, in comparision with other state-of-the-art baseline algorithms. We expect this work will shed a new light on not only protein engineering but also on solving combinatorial problems using data-driven methods.
1 March 2023 in Machine Learning: Science and Technology
In this paper, we propose Multiresolution Equivariant Graph Variational Autoencoders (MGVAE), the first hierarchical generative model to learn and generate graphs in a multiresolution and equivariant manner. At each resolution level, MGVAE employs higher order message passing to encode the graph while learning to partition it into mutually exclusive clusters and coarsening into a lower resolution that eventually creates a hierarchy of latent distributions. MGVAE then constructs a hierarchical generative model to variationally decode into a hierarchy of coarsened graphs. Importantly, our proposed framework is end-to-end permutation equivariant with respect to node ordering. MGVAE achieves competitive results with several generative tasks including general graph generation, molecular generation, unsupervised molecular representation learning to predict molecular properties, link prediction on citation graphs, and graph-based image generation. Our implementation is available at https://github.com/HyTruongSon/MGVAE.
9 January 2025 in Biology Methods and Protocols
Drug repurposing presents a promising strategy to accelerate drug discovery by identifying new therapeutic uses for existing compounds, particularly for diseases with limited or no effective treatment options. We introduce DrugPipe, a ‘Generative AI-Assisted Virtual Screening Pipeline’ developed within the target-centric paradigm of drug repurposing, which aims to discover new indications by identifying compounds that interact with a specific protein target. ‘DrugPipe’ integrates generative modeling, binding pocket prediction, and similarity-based retrieval from drug databases to enable a scalable and generalizable in silico repurposing workflow. It supports blind virtual screening for any protein target without requiring prior structural or functional annotations, making it especially suited for novel or understudied targets and emerging health threats. By efficiently generating candidate ligands and rapidly retrieving structurally similar approved drugs, ‘DrugPipe’ accelerates the identification and prioritization of repurposable compounds. In comparative evaluations, it achieves hit rate performance comparable to QVina-W, a widely used blind docking tool, while significantly reducing computational time, highlighting its practical value for large-scale virtual screening and data-scarce repurposing scenarios. The full implementation and evaluation details are available at https://github.com/HySonLab/DrugPipe.