OrgNet+, a conformational ensemble-aware and orientation-gnostic framework that explicitly incorporates protein structure flexibility during training, is introduced, which substantially reduces intra-ensemble prediction variance while simultaneously improving predictive accuracy.
Abstract
Abstract Motivation Predicting the effect of single-point mutations on protein stability is a central problem in molecular biology and protein engineering. Recent structure-based deep learning methods, particularly 3D convolutional neural networks (3D CNNs), have achieved strong predictive performance by leveraging high-resolution protein structures. However, proteins exist as heterogeneous conformational ensembles rather than single static structures, and the impact of conformational flexibility on structure-based ΔΔG predictors remains poorly characterized. Consequently, current models may yield unstable or even contradictory predictions when evaluated across alternative, yet equally plausible, conformations of the same protein. Results We introduce OrgNet+, a conformational ensemble-aware and orientation-gnostic framework that explicitly incorporates protein structure flexibility during training. OrgNet+ is trained on augmented datasets comprising diverse conformational ensembles generated using a comprehensive set of molecular modelling methods: normal mode analysis, molecular dynamics, Monte-Carlo simulations, and a generative deep learning model. Across all ensemble types, OrgNet+ substantially reduces intra-ensemble prediction variance while simultaneously improving predictive accuracy. The improved performance extends to standard single-reference-structure benchmarks, even though OrgNet+ was trained exclusively on conformational ensembles and never exposed to the reference experimental structures. Availability and implementation OrgNet+ is available at https://github.com/i-Molecule/OrgNet.
Accurately predicting mutation-induced protein stability changes remains a central challenge in structural bioinformatics. Existing methods exhibit a strong bias toward destabilizing mutations, leading to limited performance for stabilizing mutations and constraining their utility in protein engineering. Here, we address this limitation through two complementary strategies: balanced dataset construction and integrative modeling. To mitigate the severe class imbalance in current stability datasets, we constructed undersampling-based balanced datasets and further evaluated reverse-mutation augmentation as a comparative strategy. Building on the rapid development of high-performing predictors, we hypothesized that integrating their outputs could exploit complementary strengths and improve predictive accuracy. Accordingly, we developed three modeling frameworks, including models based on handcrafted features, models using embedding representations extracted from ProteinMPNN, and ensemble models integrating a diverse set of state-of-the-art predictors. Across multiple independent test sets, ensemble models consistently outperformed individual approaches, with particularly pronounced gains in identifying stabilizing mutations. These findings demonstrate that combining undersampling-based balanced data construction with systematic predictor integration provides an effective and practical strategy for achieving more balanced and accurate protein stability prediction, and offers a useful framework for identifying stabilizing mutations in protein engineering and related applications. StaMutAble is freely available at: https://github.com/minghuilab/StaMutAble.
Yang Liu, Jian Zhang, Minghui Li· Protein Science· 0 citations
Abstract Motivation Protein folding stability is a key determinant for understanding protein dynamics, including molecular function, pathogenicity, and protein engineering. Yet, accurate prediction of protein stability remains challenging due to high variability in available data, particularly when only sequence information is available and structural knowledge is limited or unavailable. In this work, we introduce LoMuS, a multi-representation-based deep learning model that predicts dataset-provided protein stability scores directly from the primary sequence. In the core of the model architecture, a fusion network integrates explicit physicochemical descriptors with low-rank adapted protein language model derived embeddings from the sequence that consistently gains across standard experimental stability benchmarks. Results We rigorously evaluate LoMuS across multiple settings, such as absolute folding stability scoring, mutation landscape stability scoring, held-out protein domains, out-of-distribution label regimes, and per-protein evaluation. LoMuS consistently outperforms sequence-only baselines, achieving an absolute performance gain of at least 10% in Spearman’s rank correlation across several benchmarks. Per-protein evaluations further demonstrate robust performance gains. Ablation analyses confirm that complementary signals from physicochemical descriptors and sequence embeddings are critical to the effectiveness of the proposed multi-representation approach. We believe LoMuS advances protein engineering research by improving the prediction and ranking of protein stability scores. Availability All codes including data preparation scripts, training and validation recipes, and experimental configurations for LoMuS are available at: https://github.com/kabir-ai2bio-lab/LoMuS.
Samuel Infante, Akash Singh, Anowarul Kabir· Bioinformatics· 0 citations
Abstract Motivation Protein dynamics are central to function, but experiments and molecular dynamics (MD) simulations remain costly, low-throughput, and difficult to compare across protocols. Scalable structure-based methods are needed to infer dynamics from static protein structures. Results We present a deep learning framework that predicts protein dynamics from 30-dimensional Gaussian integral (GI) descriptors of Cα backbone topology. Using 1374 ATLAS protein chains with MD-derived RMSF, GI stratified proteins into fold-relevant clusters enriched for secondary structure, sequence homology, and ECOD families. An attention-based 1D-CNN classified flexible versus non-flexible proteins with test AUC = 0.772 and separated slow-mode– from fast-mode–dominated dynamics with AUC = 0.91. Regression models recovered mean RMSF (Pearson r = 0.72; R² = 0.46) and slow-mode RMSF more accurately (Pearson r = 0.83; R² = 0.62), supporting rapid inference of flexibility and collective-motion bias. Availability and implementation Code and data are available on GitHub at: https://github.com/fvilicich/gaussian_integral/blob/main/gaussian_integral_classification.ipynb.
F. Vilicich, Nicolás Bottino, Zhaoqian Su et al.· Bioinform.· 0 citations
HyBind-NN is developed, a multimodal graph neural network that integrates protein language models (PLMs) with 3D structural and dynamic datasets to predict protein–protein and protein–peptide affinity, and it is demonstrated that combining ESM-2 sequence embeddings with precise 3D Voronoi spatial geometry enables accurate affinity predictions across diverse structural datasets.
E. A. Bogdanova, A. Chernukhin, Alexey K. Shaytan· International Journal of Mol...· 0 citations
Prediction of protein stability change caused by amino acid substitutions or indels (insertions/deletions) is crucial for protein engineering. While current models excel at single-point substitutions, they struggle with multi-point mutations and indels due to simplistic additivity assumptions and the inability to model backbone conformational changes. To address these limitations, we introduce UniStab, an end-to-end framework for predicting stability changes across all mutation types. By leveraging the implicit geometric reasoning of a pre-trained folding model, UniStab effectively captures non-additive epistatic interactions and local backbone rearrangements without the prohibitive cost of explicit structure generation. Evaluated on a comprehensive benchmark, UniStab demonstrates state-of-the-art performance, particularly in the challenging scenarios of multi-point mutations and indels. Beyond predictive accuracy, UniStab provides interpretable structural insights and effectively guides the design of stabilized variants, facilitating its potential utility in rational protein engineering.
Hong Tan, Shenggeng Lin, Yi Xiong· Chemical Science· 0 citations
Accurate protein structure prediction is fundamental to structural biology because protein structure underlies molecular function and provides a basis for mechanistic interpretation. Recent advances in deep learning have transformed the field from multiple sequence alignment (MSA)-driven monomer folding into broader frameworks capable of modeling protein complexes and increasingly heterogeneous molecular systems. Existing reviews have summarized this progress from the perspectives of representative models, application domains, and protein design. Building on these efforts, this review focuses on the methodological evolution of the field itself. It examines recent developments through three closely related dimensions: representations and data, architectures and learning strategies, and confidence and evaluation. Within this perspective, the field is organized into four methodological phases and three cross-cutting transitions: from explicit evolutionary coupling features and early contact prediction to learned sequence representations in AlphaFold2, RoseTTAFold, and ESMFold; from protein-only monomer folding to increasingly integrated modeling of heterogeneous molecular systems in AlphaFold-Multimer, RoseTTAFoldNA, and AlphaFold3; and, more recently, from prediction-oriented structure inference to design-oriented generative modeling in RFdiffusion and related frameworks. This framework provides a clearer understanding of how methodological shifts have shaped the capabilities, limitations, and practical roles of recent models.
Wengan He, Yongsheng Luo, Lihong Jiang et al.· 0 citations