FoldKit is introduced, a Python package for efficient storage and analysis of large-scale AF3 co-folding results that reduces storage requirements and facilitating programmatic access to relevant outputs, which facilitates large-scale computational studies of biomolecular interactions.
Abstract
AlphaFold 3 (AF3) enables structure prediction of biomolecular complexes through co-folding multiple interacting molecules, making it increasingly useful for de novo protein design and for large-scale studies of protein-protein, protein-peptide, and other biomolecular interactions. However, systematic co-folding experiments can produce large volumes of output data, particularly when multiple random seeds and samples are generated for each input complex. We introduce FoldKit, a Python package for efficient storage and analysis of large-scale AF3 co-folding results. FoldKit converts raw AF3 outputs into a compact, structured representation while preserving the metadata needed for downstream analysis. The FoldKit Python library provides convenient programmatic access to global, single chain, and interface confidence metrics such as pLDDT, pTM, ipTM, ipAE, and ipSAE, as well as an ensemble-level interface for accessing and aggregating these metrics for a single input across multiple seeds and samples. We benchmark FoldKit on three types of AF3 co-folding datasets: (i) a protein design campaign with 2 chains per input, (ii) a TCR-pMHC dataset with 4 chains per input, and (iii) a pooled-AF3 protein-protein interaction dataset with up to 22 chains per input. We find that FoldKit reduces storage requirements by approximately 5-15-fold compared to native AF3 outputs, depending on dataset composition, while maintaining direct programmatic access to individual predictions, ensembles, and confidence metrics. By reducing storage requirements and facilitating programmatic access to relevant outputs, FoldKit facilitates large-scale computational studies of biomolecular interactions. FoldKit is available from PyPI and can be installed using pip.
The Atlas model family is presented, an open and trainable system spanning protein language modeling, monomer folding, and protein-complex prediction, and the training code and data, stage checkpoints, and model weights under the MIT License are released to provide a foundation for advancing PLM-based protein structure prediction.
Seonghwan Seo, Hyeongwoo Kim, Seokhyun Moon et al.· bioRxiv· 0 citations
COACH-D 2.0 is introduced, a substantially enhanced template-based method for predicting protein-ligand binding sites and features three key advances: integration of multimeric templates from Q-BioLiP into the authors' in-house library, a new multimeric structure processing module enabling binding site prediction for protein complexes, and an efficient template screening strategy that significantly boosts both prediction speed and accuracy.
Xiao-Yu An, Hong Wei, Wenkai Wang et al.· Genomics, Proteomics & Bioin...· 0 citations
Accurate annotation of RNA base-pairing interactions is essential for structural analysis, benchmarking, and data-driven RNA structure prediction. Several tools can extract RNA interactions from three-dimensional coordinates, but their outputs are heterogeneous and may disagree, particularly for non-canonical base pairs. We present EXTRARNAS, a Java-based framework for automated, reproducible, and user-friendly large-scale extraction of RNA structural annotations with multiple tools. EXTRARNAS processes batches of RNA structures specified by PDB identifier and chain, or provided as local PDB files, executes annotation tools through a Docker-based environment, and parses tool-specific outputs using ANTLR4-based grammars. For each structure–tool pair, the framework generates standard BPSEQ files for canonical cis Watson–Crick interactions and introduces BPSEQE, a standardized text format for representing the extended secondary structure, preserving canonical, non-canonical, and multiple interactions per nucleotide. The current prototype supports RNAView, MC-Annotate, and RNAPolis Annotator. We demonstrate EXTRARNAS on eight RNA structures containing triple-helix motifs, comparing extracted canonical pairs against curated BPSEQ references and evaluating the recovery of manually validated Hoogsteen interactions. The results show consistent differences among tools, especially for non-canonical interactions, highlighting the need for standardized representations such as BPSEQE to support reproducible comparison and future consensus-based annotation.
Federico Di Petta, Piermichele Rosati, Piero Hierro Canchari et al.· bioRxiv· 0 citations
SoupFold is introduced, which improves co-folding predictions by learning simple mappings between the representation spaces of co-folding models, and achieves state-of-the-art performance on both protein-protein and protein-ligand structure prediction.
Hyosoon Jang, Taewon Kim, Sungsoo Ahn· 0 citations
MultiStructRNA enables seamless switching between prediction methods without requiring workflow changes and supports both in-notebook and exportable visualizations, and supports high-throughput analyses and simplifies comparison across methods while standardizing downstream feature extraction.
Protein language models (PLMs) provide powerful representations of protein sequence, but their utility for proteome-scale binding-site retrieval remains unclear. Here, we present PocketScope, a training-free framework that represents cavity-lining residues using frozen ESM-C 600M embeddings and retrieves related binding sites through exhaustive lateinteraction MaxSim, without pooling or approximate nearest-neighbor search. PocketScope identified 153,805 cavities across 37,682 proteins in the AlphaFold human proteome and recovered documented drug off-targets across a curated set of pharmacological pairs. On the ProSPECCTs benchmark, PocketScope ranks 1st of 23 methods by mean rank across the ten collections. PocketScope provides a practical framework for proteome-scale off-target prediction. PocketScope is open source and also freely available as a web server at https://www.bhargavaresearch.org/pocketscope.
Keshav Mohan, Yash Bhargava· bioRxiv· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduAug 27, 2026
A new machine-learning framework aims to improve the success rate of computational protein design while moving away from results that reproduce sequences found in nature.
