{"count":10,"items":[{"article":null,"cancer_slugs":["pan-cancer"],"category":"protein-structure","confidence":"high","developer":"Google DeepMind","hf":null,"kind":"foundation","license":"Apache-2.0 (code) / CC-BY-4.0 (parameters)","links":{"demo":"https://alphafold.ebi.ac.uk/","docs":null,"doi":"10.1038/s41586-021-03819-2","github":"https://github.com/google-deepmind/alphafold","huggingface":null,"paper":"https://www.nature.com/articles/s41586-021-03819-2","pmid":null},"modalities":["protein-sequence"],"name":"AlphaFold 2","openness":"open-weights","regulatory_status":"not-applicable","release_date":"2021-07-15","settings":["basic-research","drug-discovery"],"slug":"alphafold","summary":"Predicts the 3D structure of a protein from its amino-acid sequence at near-experimental accuracy; the AlphaFold Protein Structure Database (with EMBL-EBI) provides predicted structures for over 200 million proteins, including cancer-relevant targets and mutants.","tasks":["protein-structure"],"url":"/ai-oncology/models/alphafold","verified_at":"2026-09-05T22:26:00.969670","version":"2"},{"article":null,"cancer_slugs":["pan-cancer"],"category":"protein-structure","confidence":"high","developer":"Google DeepMind / Isomorphic Labs","hf":null,"kind":"foundation","license":"CC-BY-NC-SA 4.0 (code); model parameters under academic non-commercial terms","links":{"demo":"https://alphafoldserver.com/","docs":null,"doi":"10.1038/s41586-024-07487-w","github":"https://github.com/google-deepmind/alphafold3","huggingface":null,"paper":"https://www.nature.com/articles/s41586-024-07487-w","pmid":null},"modalities":["protein-sequence","molecules","genomics"],"name":"AlphaFold 3","openness":"gated-weights","regulatory_status":"not-applicable","release_date":"2024-05-08","settings":["basic-research","drug-discovery"],"slug":"alphafold-3","summary":"Diffusion-based successor that predicts joint structures of proteins with DNA, RNA, ligands, ions and modified residues \u2014 the interaction types that matter for drug design and for understanding oncogenic complexes.","tasks":["protein-structure","drug-discovery"],"url":"/ai-oncology/models/alphafold-3","verified_at":"2026-09-05T22:26:00.993178","version":"3"},{"article":null,"cancer_slugs":["pan-cancer"],"category":"protein-structure","confidence":"high","developer":"Meta AI (FAIR)","hf":{"downloads":1303383,"fetched_at":"2026-09-09T21:33:09Z","gated":false,"last_modified":"2023-03-21","library":"transformers","license":"mit","likes":89,"pipeline_tag":"fill-mask"},"kind":"foundation","license":"MIT","links":{"demo":null,"docs":null,"doi":"10.1126/science.ade2574","github":"https://github.com/facebookresearch/esm","huggingface":"https://huggingface.co/facebook/esm2_t33_650M_UR50D","paper":"https://www.science.org/doi/10.1126/science.ade2574","pmid":null},"modalities":["protein-sequence"],"name":"ESM-2 / ESMFold","openness":"open-weights","regulatory_status":"not-applicable","release_date":"2022-11-01","settings":["basic-research","drug-discovery"],"slug":"esm-2","summary":"Protein language models from 8 million to 15 billion parameters trained on UniRef sequences; embeddings power variant-effect and function prediction, and ESMFold predicts structure directly from a single sequence without MSAs.","tasks":["feature-extraction","variant-effect","protein-structure"],"url":"/ai-oncology/models/esm-2","verified_at":"2026-09-05T22:26:01.022006","version":null},{"article":"/blog/modele-prezentacji-antygenu","cancer_slugs":["pan-cancer"],"category":"immunopeptidomics","confidence":"high","developer":"Broad Institute / Dana-Farber Cancer Institute (Keskin, Wu, Carr labs)","hf":null,"kind":"task-model","license":"free web server; academic use","links":{"demo":"http://hlathena.tools/","docs":null,"doi":"10.1038/s41587-019-0322-9","github":null,"huggingface":null,"paper":"https://pmc.ncbi.nlm.nih.gov/articles/PMC7008090/","pmid":null},"modalities":["protein-sequence","immunopeptidomics","transcriptomics"],"name":"HLAthena","openness":"api-only","regulatory_status":"research-only","release_date":"2020-01-13","settings":["basic-research","vaccine-design","immunotherapy"],"slug":"hlathena","summary":"Presentation predictor trained on the mono-allelic peptidome of 95 cell lines \u2014 186,464 peptides across 95 HLA alleles, fifteen of which had no described motif before \u2014 which is the dataset that changed this field more than any architectural idea.","tasks":["antigen-presentation","peptide-mhc-binding","neoantigen-prioritisation"],"url":"/ai-oncology/models/hlathena","verified_at":"2026-09-05T22:26:00.948597","version":null},{"article":"/blog/modele-prezentacji-antygenu","cancer_slugs":["pan-cancer"],"category":"immunopeptidomics","confidence":"high","developer":"O'Donnell, Rubinsteyn and Laserson (Mount Sinai / OpenVax)","hf":null,"kind":"task-model","license":"apache-2.0","links":{"demo":null,"docs":null,"doi":"10.1016/j.cels.2020.06.010","github":"https://github.com/openvax/mhcflurry","huggingface":null,"paper":"https://www.cell.com/cell-systems/fulltext/S2405-4712(20)30331-8","pmid":null},"modalities":["protein-sequence","immunopeptidomics"],"name":"MHCflurry 2.0","openness":"open-weights","regulatory_status":"research-only","release_date":"2020-07-08","settings":["basic-research","vaccine-design","immunotherapy"],"slug":"mhcflurry","summary":"Open-source pan-allele presentation predictor whose distinctive idea is a separate antigen-processing model: it reads the peptide together with fifteen amino acids of flanking sequence on each side, because the proteasome cut depends on what lies around the cut site.","tasks":["antigen-presentation","peptide-mhc-binding","neoantigen-prioritisation"],"url":"/ai-oncology/models/mhcflurry","verified_at":"2026-09-05T22:26:00.839223","version":"2.0"},{"article":"/blog/modele-prezentacji-antygenu","cancer_slugs":["pan-cancer"],"category":"immunopeptidomics","confidence":"medium","developer":"Karchin lab, Johns Hopkins University","hf":null,"kind":"task-model","license":"apache-2.0","links":{"demo":null,"docs":null,"doi":"10.1158/2326-6066.CIR-19-0464","github":"https://github.com/KarchinLab/mhcnuggets","huggingface":null,"paper":"https://pmc.ncbi.nlm.nih.gov/articles/PMC7056596/","pmid":null},"modalities":["protein-sequence","immunopeptidomics"],"name":"MHCnuggets","openness":"open-weights","regulatory_status":"research-only","release_date":"2020-03-01","settings":["basic-research","vaccine-design","immunotherapy"],"slug":"mhcnuggets","summary":"Allele-specific LSTM networks \u2014 one per allele, 148 for class I \u2014 that read peptides letter by letter, so no alignment or padding is needed, and that are trained by transfer learning from the data-richest allele.","tasks":["peptide-mhc-binding","antigen-presentation","neoantigen-prioritisation"],"url":"/ai-oncology/models/mhcnuggets","verified_at":"2026-09-05T22:26:00.909264","version":"2.x"},{"article":"/blog/modele-prezentacji-antygenu","cancer_slugs":["pan-cancer"],"category":"immunopeptidomics","confidence":"medium","developer":"Gfeller lab, Ludwig Institute for Cancer Research / University of Lausanne","hf":null,"kind":"task-model","license":"free for academic use (see repository terms)","links":{"demo":null,"docs":null,"doi":"10.1186/s13073-025-01449-1","github":"https://github.com/GfellerLab/MixMHCpred","huggingface":null,"paper":"https://pmc.ncbi.nlm.nih.gov/articles/PMC11927126/","pmid":null},"modalities":["protein-sequence","immunopeptidomics"],"name":"MixMHCpred 3.0","openness":"open-weights","regulatory_status":"research-only","release_date":"2025-03-19","settings":["basic-research","vaccine-design","immunotherapy"],"slug":"mixmhcpred","summary":"Position-weight-matrix predictor built on the observation that class I ligands show almost no dependence between positions; version 3.0 closes the circle by having a neural network predict the matrix from the same 34 groove residues.","tasks":["antigen-presentation","peptide-mhc-binding","neoantigen-prioritisation"],"url":"/ai-oncology/models/mixmhcpred","verified_at":"2026-09-05T22:26:00.871048","version":"3.0"},{"article":"/blog/modele-prezentacji-antygenu","cancer_slugs":["pan-cancer"],"category":"immunopeptidomics","confidence":"medium","developer":"Health Tech, Technical University of Denmark (Nielsen lab)","hf":null,"kind":"task-model","license":"free for academic use; commercial licence required (DTU)","links":{"demo":null,"docs":"https://services.healthtech.dtu.dk/services/NetMHCIIpan-4.0/","doi":"10.1093/nar/gkaa379","github":null,"huggingface":null,"paper":"https://academic.oup.com/nar/article/48/W1/W449/5837056","pmid":null},"modalities":["protein-sequence","immunopeptidomics"],"name":"NetMHCIIpan-4.0","openness":"api-only","regulatory_status":"research-only","release_date":"2020-05-22","settings":["basic-research","vaccine-design","immunotherapy"],"slug":"netmhciipan","summary":"The class II counterpart of NetMHCpan, released in the same paper: predicts presentation by HLA-DR, -DQ and -DP, whose groove is open at both ends, so peptides are longer and the binding core has to be found inside a longer sequence.","tasks":["antigen-presentation","peptide-mhc-binding","neoantigen-prioritisation"],"url":"/ai-oncology/models/netmhciipan","verified_at":"2026-09-05T22:26:00.814125","version":"4.0"},{"article":"/blog/modele-prezentacji-antygenu","cancer_slugs":["pan-cancer"],"category":"immunopeptidomics","confidence":"high","developer":"Health Tech, Technical University of Denmark (Nielsen lab)","hf":null,"kind":"task-model","license":"free for academic use; commercial licence required (DTU)","links":{"demo":null,"docs":"https://services.healthtech.dtu.dk/services/NetMHCpan-4.1/","doi":"10.1093/nar/gkaa379","github":null,"huggingface":null,"paper":"https://academic.oup.com/nar/article/48/W1/W449/5837056","pmid":null},"modalities":["protein-sequence","immunopeptidomics"],"name":"NetMHCpan-4.1","openness":"api-only","regulatory_status":"research-only","release_date":"2020-05-22","settings":["basic-research","vaccine-design","immunotherapy"],"slug":"netmhcpan","summary":"The reference pan-allele predictor of MHC class I antigen presentation: one small neural network covers more than 11,000 MHC molecules because the groove itself is part of the input \u2014 a 34-residue pseudosequence next to the peptide.","tasks":["antigen-presentation","peptide-mhc-binding","neoantigen-prioritisation"],"url":"/ai-oncology/models/netmhcpan","verified_at":"2026-09-05T22:26:00.703103","version":"4.1"},{"article":null,"cancer_slugs":["pan-cancer"],"category":"genomics","confidence":"high","developer":"Google DeepMind","hf":null,"kind":"task-model","license":"Apache-2.0 (code); predictions CC-BY-NC-SA 4.0","links":{"demo":null,"docs":null,"doi":"10.1126/science.adg7492","github":"https://github.com/google-deepmind/alphamissense","huggingface":null,"paper":"https://www.science.org/doi/10.1126/science.adg7492","pmid":null},"modalities":["genomics","protein-sequence"],"name":"AlphaMissense","openness":"open-weights","regulatory_status":"research-only","release_date":"2023-09-19","settings":["diagnosis","basic-research"],"slug":"alphamissense","summary":"Classifies the pathogenicity of every possible single amino-acid substitution in the human proteome \u2014 71 million missense variants \u2014 by fine-tuning AlphaFold on population variant frequencies; a resource for interpreting variants of uncertain significance.","tasks":["variant-effect"],"url":"/ai-oncology/models/alphamissense","verified_at":"2026-09-07T15:15:28.911174","version":null}],"standard":"https://cancer3.ai/docs/ai-model-card","vocab":{"confidence":["high","medium","low"],"dataset_access":["open","registration","dua","controlled","commercial"],"dataset_kinds":["dataset","registry","benchmark","biobank","corpus"],"formats":["SVS","TIFF","NDPI","DICOM","NIfTI","PNG","JPEG","CSV","TSV","JSON","Parquet","FASTQ","BAM","VCF","MAF","h5ad","PDB","mmCIF","text","other"],"kinds":["foundation","task-model","tool","framework","product"],"link_roles":["pretraining","training","finetuning","evaluation","benchmark"],"modalities":["histopathology","radiology-ct","radiology-mri","radiology-xray","mammography","pet","ultrasound","dermoscopy","endoscopy","genomics","transcriptomics","single-cell","proteomics","protein-sequence","molecules","clinical-text","literature","ehr","multimodal","immunopeptidomics"],"need_status":["open","in-progress","solved"],"openness":["open-weights","gated-weights","api-only","closed"],"regulatory":["research-only","fda-cleared","fda-de-novo","fda-pma","ce-marked","ce-ivdr","not-applicable"],"settings":["basic-research","screening","diagnosis","prognosis","treatment-planning","treatment-response","drug-discovery","clinical-trials","education","vaccine-design","immunotherapy"],"tasks":["feature-extraction","classification","segmentation","detection","screening","risk-prediction","prognosis","treatment-response","survival-analysis","image-text-retrieval","report-generation","question-answering","information-extraction","clinical-trial-matching","protein-structure","variant-effect","gene-expression","single-cell","drug-discovery","drug-repurposing","peptide-mhc-binding","antigen-presentation","neoantigen-prioritisation"]}}