Neurosnap
No-code and API inference for protein folding, design, and docking — run AlphaFold 2, RFdiffusion, ProteinMPNN, and RoseTTAFold All-Atom online.
Overview
Neurosnap is a hosted directory of protein modeling services that lets you run folding, design, and docking tools straight from the browser or through an API. It removes the setup burden that usually stands between a sequence and a prediction: there is no environment to build, no GPU to rent, and no weights to download. For anyone searching for an online AlphaFold service or a place to run protein design without local infrastructure, Neurosnap organizes the field's core tools into individually addressable services.
What you can run on Neurosnap
The service catalog centers on protein structure and design. AlphaFold 2 predicts protein structure from sequence, and RoseTTAFold All-Atom extends structure prediction across proteins, nucleic acids, small molecules, and covalent modifications for mixed biomolecular assemblies. On the design side, RFdiffusion generates novel protein backbones for de novo design and scaffolding tasks, while ProteinMPNN performs inverse folding to produce sequences that fold to a chosen structure. This mix spans the essential steps of a structure-and-design workflow — folding, all-atom complex modeling, backbone generation, and sequence design — and supports docking-oriented analysis.
Running inference on Neurosnap
Each tool is delivered as hosted inference: a stable per-service page in the web directory for interactive, no-code runs, plus a REST API for programmatic and batch submission. That dual access makes it approachable for wet-lab biologists who want a point-and-click job and equally usable by developers scripting pipelines or wiring the endpoints into automated agents. The platform is focused on running models rather than distributing weights or offering fine-tuning, so it fits users who need quick, reliable access to established protein tools and want to move between folding and design methods in one place.
Run inference on Neurosnap (4)
Protein structure prediction model that folds amino acid sequences into 3D structures with atomic accuracy, scoring a median GDT of 92.4 at CASP14.
Message passing neural network for fixed-backbone protein sequence design. Achieves 52.4% native sequence recovery, far surpassing Rosetta's 32.9%.
De novo protein design diffusion model that generates backbone structures conditioned on binding targets, symmetry constraints, and functional motifs.
Deep network that predicts structures of full biological assemblies: proteins, nucleic acids, small molecules, metals, and covalent modifications.