BioMetAll v2.0: Introducing Scores, Metal Discrimination, and Side-Chain Descriptors for Predicting Metal-Binding Sites in Proteins

Capabilities, specificity gaps and training-data dependence of AlphaFold3 across diverse application areas

It is found that, while AF3 can perform well in favourable settings, this performance is uneven across applications and its predictions and use of confidence metrics will depend strongly on the specific application area and must be interpreted with respect to training-set overlap.

Nesso-1: Accelerating Open-Source Binding Affinity Predictions

In this technical report, we introduce Nesso-1, a coarse-grained cofolding framework for binding- affinity prediction. Nesso-1 requires ∼ 1 second per prediction on a single GPU. This offers more than one order of magnitude speed-up over the leading open-source baseline, Boltz-2, which significantly expands the regions of chemical space that can be explored during high-throughput virtual screening. Importantly, Nesso-1 matches or surpasses the accuracy of Boltz-2 over the same benchmarks adopted in their study—which we show reflect in-distribution scenarios—as well as over more challenging out-of-distribution data encompassing the OpenBind affinity benchmark and 25 internal biochemical assays. Notably, Nesso-1 maintains robust predictive accuracy even on assays with extremely low similarity to the training data. Moreover, we highlight examples where Nesso-1 demonstrates meaningful selectivity, separating the binding affinities of identical compounds between on-targets and related off-targets. Nonetheless, zero-shot generalization to real- world medicinal chemistry remains an inherently challenging task; consequently, we acknowledge specific assays where the model’s performance is limited. We open-source Nesso-1: code and weights are available at https://github.com/recursionpharma/nesso

Deep-learning predictions of biomolecular structures : persistent limitations and new horizons extended by explicit ion addition

COACH-D 2.0: A Server for Template-based Modeling of Protein-ligand Interactions.

Protein structure prediction has been transformed by AlphaFold2 and related systems; however, accurately modeling protein-ligand interactions remains a major challenge. Template-based prediction using homologous structures remains a powerful strategy. Here, we introduce COACH-D 2.0, a substantially enhanced template-based method for predicting protein-ligand binding sites. This upgrade features three key advances: (1) integration of multimeric templates from Q-BioLiP into our in-house library, (2) a new multimeric structure processing module enabling binding site prediction for protein complexes, and (3) an efficient template screening strategy that significantly boosts both prediction speed and accuracy. Evaluations against the previous version and leading methods on three benchmark datasets demonstrate the superior performance of COACH-D 2.0. The server is freely accessible at https://yanglab.qd.sdu.edu.cn/COACH-D/.

Application of vision transformers to protein-ligand affinity prediction

Despite challenges related to data sparsity and conformational variability, ViTs show strong performance and high robustness in structure-based affinity prediction tasks, underscore their effectiveness in learning spatial patterns and suggest broader applicability to related tasks, such as protein-protein or protein-nucleic acid interaction modeling.

Mavchen-1: A Conformational Ensemble Platform for Protein–Ligand Pose Prediction That Substantially Outperforms Static Structure Prediction in a Category-Stratified Benchmark

A category-stratified, statistically powered benchmark comparing pose prediction from receptor conformational ensembles against AlphaFold2, used as a matched static-structure baseline, across 29 protein–ligand systems spanning cryptic-pocket, induced-fit, water-mediated, and autoimmune-indication target classes is presented.