Corollary

Research

  • New investigation
  • Papers

Library

  • Catalog

Account

  • Jobs
  • Billing
  • Developers
SettingsResearch demo · not for clinical or commercial use
Corollary
  1. Catalog

Library

Find the model or dataset for the question you have.

Ask in plain language and we will work out where the answer lives — or filter the index yourself. Almost everything here runs on our servers, from this page, with no setup.

Models
300
Datasets
248
Organisations
131

Labs, institutes and companies

Ready to run
21

Including 6 that run free and instantly

Looking for a particular value, not a particular dataset?Search schemas, rows and column statistics inside the data itself.

Things that work

Filtering

23 entries in Biotechnology

Showing 23 of 23

Ready to run

3
ReadyModel

ProGen2

Salesforce Research

Autoregressive protein generation up to 6.4B parameters, with an antibody variant.

sequence generation
ReadyModel

ProteinMPNN

Baker Lab

Inverse folding: given a backbone, design sequences that actually fold onto it.

inverse folding
ReadyModel

RFDiffusion

Baker Lab / Institute for Protein Design

De novo backbone generation by denoising diffusion — the binder-design workhorse.

backbone generation

Everything else indexed

20
AbLang2Oxford Protein Informatics GroupAntibody-specific language model trained on the Observed Antibody Space.sequence embeddingOn demand
ANARCIOxford Protein Informatics GroupAntibody numbering and region annotation — Chothia, IMGT, Kabat.annotationOn demand
AntiFoldHummer et al., DTUInverse folding specialised for antibody backbones and CDR loops.inverse foldingOn demand
ESM-IF1Meta FAIRLanguage-model-informed inverse folding — an alternative view to ProteinMPNN.inverse foldingOn demand
ImmuneBuilderOxford Protein Informatics GroupSub-second structure prediction for antibodies, nanobodies and TCRs.structure predictionOn demand
MAMMAL solubilityIBM ResearchBinary protein-solubility classification from amino-acid sequence.protein solubilityOn demand
SapiensAstraZeneca / Prihoda et al.Germline-guided antibody humanization, validated on clinical-stage molecules.humanizationOn demand
SaProtSu et al., Westlake UniversityStructure-aware protein LM using combined sequence + structure tokens.sequence embeddingOn demand
GDPa1ginkgo-datapointsAntibody developability dataset with biophysical assay data for 242 antibodies across 9 assays.Antibody DevelopabilityExplore
GDPx1ginkgo-datapointsDRUG-seq functional genomics dataset with chemical perturbation experiments in A549 cells.Functional GenomicsExplore
GDPx2ginkgo-datapointsDRUG-seq transcriptomic profiling across 4 primary human cell types with 85 compounds.Functional GenomicsExplore
GDPx3ginkgo-datapointsHigh-content Cell Painting imaging dataset for AI/ML model training in drug discovery.Cell ImagingExplore
GDPx4ginkgo-datapointsDRUG-seq transcriptomic profiling in engineered HEK293 cells with inducible gene overexpression, enabling systematic study of gene-drug interactions.Functional GenomicsExplore
img_virus_plasmidwanglabCombined IMG/VR (uncultivated virus genomes) and IMG/PR (plasmids from genomes and metagenomes) catalog with rich functional, taxonomic, and ecological metadata.Microbial GenomicsExplore
BindCraftOvchinnikov Lab / Baker LabAF2-guided hallucination that co-folds target and binder — ~10 screens to a hit.binder designBrowse
DPLM-2Nanjing University / ByteDanceDiffusion protein language model generating sequence and structure jointly.sequence generationBrowse
EVOLVEproHarvard / Broad InstituteFew-shot active learning for directed evolution — 24 mutants per round.directed evolutionBrowse
IgFoldRuffolo et al., Johns HopkinsAntibody structure prediction ~1000× faster than AlphaFold2.structure predictionBrowse
ParagraphDeane Lab / OxfordParatope and epitope prediction for humanization and immunogenicity work.epitope predictionBrowse
RudolfV 2AignosticsAignostics recently announced RudolfV 2, the best-performing openly available pathology foundation model to date. The model is built on a ViT-g architecture with approximately 1.1 billion parameters, and includes a dedicated post-training stage that to our knowledge is a first for pathology foundation models. We further distilled RudolfV 2 into two lightweight versions, RudolfV 2-B (ViT-B, 86M parPathology Foundation ModelBrowse