Skip to main content

    Structure Prediction overview

    Fetch a known structure by UniProt ID or predict one from sequence, read its confidence metrics before you trust it, then protonate it for electrostatics.

    sectionstructure-prediction

    Before you can dock a molecule, find a binding site, or reason about a mutation, you need a 3D structure. This section gets you one — fetch a known structure or predict a novel or engineered sequence — plus the tools to judge and analyze it. Structure is the entry point for most target-side workflows.

    Availability

    These tools are metered by credits and gated by your plan — plans set your monthly credits, available model families, and which modal-app tools you can run. See Plans & tiers.

    What you can achieve here

    • Get a 3D structure — fetch a known one, or predict a novel/engineered sequence
    • Find and rank druggable binding pockets
    • Assess model confidence before you trust it downstream
    • Compare a sequence against known structures and homologs

    For which tool fits each, see What you can achieve.

    Tools in this section

    ToolWhat it does
    ESMFold on the platformPredict protein structures with ESMFold and interpret confidence metrics.
    AlphaFold on the platformFetch pre-computed AlphaFold structures from the EBI AlphaFold Database by UniProt ID.
    ESM-3 on the platformEmbed a sequence with ESM-3 for alignment-free similarity, zero-shot function prediction and clustering — per-residue and per-sequence vectors, not a…
    PDB Search & Download on the platformSearch and retrieve protein structures from the RCSB Protein Data Bank.
    PDB2PQRAssign protonation states and prepare proteins for electrostatics calculations.
    BLASTAlign a nucleotide or protein query against NCBI databases and score the hits, locally or remotely from a job, then fold the top hit with ESMFold.

    Recipes using these tools

    Was this page helpful?

    On this page

    On this page

    Loading...