Jul 2026· Practice and Experience in Advanced Research Computing· 0 citations· 10 references
Computer Science
TL;DR
The VizFold plugin is described, a modular framework that can be extended toward end-to-end composable pipelines and demonstrated feasibility through standardized hook-based tracing for ESMFold and Boltz-2, archive validation, and reproducible deployment on an HPC cluster using managed caches, modules, quotas, and Slurm workflows.
Abstract
Current AI protein structure prediction models involve multi-stage processing that combines deep neural networks with bioinformatics tools such as multiple sequence alignment (MSA). Researchers increasingly rely on intermediate or penultimate-layer activations from these models for downstream tasks including contact prediction, binding-site identification, and model interpretability. We describe the VizFold plugin, a modular framework that can be extended toward end-to-end composable pipelines. We demonstrate feasibility through standardized hook-based tracing for ESMFold and Boltz-2, archive validation, and reproducible deployment on an HPC cluster using managed caches, modules, quotas, and Slurm workflows. The framework extracts attention maps from user-selected layers and exports intermediate representations in a backend-specific run bundle (Boltz) or a canonical archive tree (ESMFold), with shared trace text conventions and explicit provenance metadata suitable for cross-model comparison. We provide step-by-step documentation for instrumentation and deployment so that other groups can reproduce or extend the pipeline on their own clusters. The framework and instrumentation code are open-source and available at https://github.com/AI2Science/vizfold-foundation.
Deep-Interact Studio is, to the authors' knowledge, the only such platform to combine fine-grained per-layer model customization with multi-model comparison and interpretability, offering a flexible and transparent alternative to fixed, single-purpose tools.
Dipayan Sarkar, K. Bardhan, Chiranjib Sarkar· bioRxiv· 0 citations
This tutorial provides a comprehensive, end-to-end workflow from raw data to deployed models,icitly designed for environmental chemists with limited prior experience in ML modeling while also providing practical guidance for other users seeking to strengthen their modeling workflows.
Kai Zhang, Yushu Cheng, Haiping Ai et al.· ACS Environmental Au· 0 citations
We introduce MultiStructRNA, a unified Python toolkit for RNA secondary structure prediction, ensemble analysis, and visualization. Although RNA secondary structure is central to RNA biology and therapeutic design, practical adoption is often hindered by fragmented tooling, incompatible input and output formats, and limited visualization support. MultiStructRNA addresses these challenges through a single high-level API that orchestrates multiple prediction algorithms, harmonizes results into a consistent schema, and provides reproducible, ensemble-aware metrics through an object model suited to both interactive notebooks and production pipelines. MultiStructRNA enables seamless switching between prediction methods without requiring workflow changes and supports both in-notebook and exportable visualizations. Designed for scalability, it supports high-throughput analyses and simplifies comparison across methods while standardizing downstream feature extraction. The current release also includes optional agent-readable workflow recipes that document dependency setup, backend-adapter conventions, SHAPE-data reconciliation, structure interpretation, and comparative sequence analyses. By integrating diverse RNA secondary structure packages within a common framework, MultiStructRNA streamlines structure analysis and facilitates its use in RNA design, optimization, and machine learning workflows.
Protein function annotation is crucial for understanding biological processes and mechanisms. Traditionally, annotations rely on sequence homology, providing valuable insights but often leaving gaps even in well-characterised organisms. With AlphaFold enabling rapid generation of protein structural models, we can now infer function from three-dimensional shape. Here, we present WASP, a pipeline leveraging structural homology to enhance protein annotation prediction at scale, providing a more comprehensive understanding of protein functions across various organisms. WASP relies on network topology for better accuracy and more robust statistical power. We show that WASP achieves superior F1 scores compared to state-of-the-art sequence-based tools when recovering hidden annotations. On 20 industrially relevant organisms, WASP retrieves annotations for 20-30% of previously uncharacterised proteins. We further demonstrate utility in genome-scale metabolic model curation, identifying native candidates for 75-100% of orphan reactions. WASP highlights how structural homology can systematically discover annotations missed by sequence-based approaches. WASP predicts protein functions from AlphaFold structures using network-based structural homology, retrieving annotations for 20-30% of uncharacterised proteins and filling metabolic model gaps by mapping 75-100% of orphan reactions.
Implementing novel features and experimental algorithms into widely adopted density functional theory (DFT) codes is frequently hindered by complex legacy architectures and the use of compiled languages such as Fortran. These production codes, while optimised for high-performance computing clusters, present significant hurdles for software development and rapid prototyping, often requiring deep expertise in the code's internal structure to modify. To address this challenge, we present a Python plugin infrastructure for the Vienna ab-initio Simulation Package (VASP) that combines computational efficiency with the flexibility of high-level scripting. Our architecture uses a C++ intermediate layer and pybind11 to expose VASP data as NumPy arrays via shared memory buffers, ensuring high performance without data duplication. We implement two categories of plugins: those that modify quantities at the end of each converged self-consistent field (SCF) cycle, such as structure and force_and_stress, and those that operate during the SCF cycle, such as local_potential and occupancies. We demonstrate the utility of our implementation through three applications, structure relaxation using the scipy library, implementing an implicit solvent model, and adding the DFT-D4 dispersion corrections. This infrastructure effectively bridges the gap between high-performance electronic structure routines and the widespread scientific Python ecosystem.
Sudarshan Vijay, M. Marsman, Georg Kresse et al.· 0 citations
Genetic testing reports coding variants far faster than they can be interpreted, and placing a variant in its biophysical context, the domain it perturbs, whether its residue is buried or exposed, whether it lies near a disulfide bond or a predicted binding pocket increasingly requires projecting it onto a three dimensional protein model. The Genomics 2 Proteins (G2P) portal unifies the gene-to-structure identifier chain with a dense, residue indexed annotation table, but its visualization layer presumes a desktop browser and is awkward at the bedside, in the classroom, or in the field, where a phone or tablet is often the only device. We present G2P Explorer, a native iOS framework that consumes the public G2P REST API on device, parses its seventy one column tab separated feature tables without loss of fidelity, and presents the result through six interlinked modules sharing a single observable view model. Beyond porting, it contributes a SwiftUI Canvas multi-track sequence renderer, an on device reconstruction of the portal's unavailable isoform alignment route, a bidirectional Swift JavaScript structural bridge that absorbs the AlphaFold model file versioning scheme, and a fault tolerant ingestion layer that parses semi structured free text and distinguishes absent annotations from zero. Each searched protein is cached on device after the first fetch, so it reopens instantly and works offline, and the embedded structural view is drawn in a reduced form suited to a small screen. Across six proteins spanning 189-1{,}863 residues, the framework sustains interactive frame times (3.1-16.6\,ms) and modest memory (16-72\,MB). G2P Explorer is an open, reproducible mobile companion to the G2P portal for hypothesis generation, teaching, and on the go variant interpretation.