---
title: "Appendix F: Model and Dataset Index – The Life Sciences AI Handbook"
source: "https://lifesciencesaihandbook.com/appendices/model-dataset-index.html"
author: "Bryan Tegomoh, MD, MPH"
last_updated: "2026-11-06"
doi: "10.5281/zenodo.22073135"
license: CC BY 4.0
generated-by: handbook-markdown-export
---

# Appendix F: Model and Dataset Index

The tables below index selected models, datasets, benchmarks, and infrastructure cited across the handbook, organized by category. Evidence labels distinguish peer-reviewed, preprint, official-program, and infrastructure claims from marketing or vendor-reported claims.

## Protein Structure Prediction

| Model / Resource | Main Use | Evidence Tier | Source |
|----|----|----|----|
| AlphaFold 2 | Single-chain protein structure prediction | Demonstrated | [Jumper et al., 2021](https://doi.org/10.1038/s41586-021-03819-2) |
| RoseTTAFold | Open AlphaFold-2-class structure prediction | Demonstrated | [Baek et al., 2021](https://doi.org/10.1126/science.abj8754) |
| ESMFold | MSA-free protein structure prediction | Demonstrated | [Lin et al., 2023](https://doi.org/10.1126/science.ade2574) |
| AlphaFold 3 | Biomolecular interaction structure prediction | Peer-reviewed; noncommercial code and weights released November 2024; source code relicensed under Apache 2.0 in June 2026 while weights remain under noncommercial terms | [Abramson et al., 2024](https://doi.org/10.1038/s41586-024-07487-w); [addendum](https://doi.org/10.1038/s41586-024-08416-7); [release terms](https://github.com/google-deepmind/alphafold3/blob/v3.0.0/WEIGHTS_TERMS_OF_USE.md); [Apache 2.0 code license](https://github.com/google-deepmind/alphafold3/blob/main/LICENSE) |
| AlphaFold-Multimer | Protein complex structure prediction | Preprint evidence | [Evans et al., 2021, preprint](https://doi.org/10.1101/2021.10.04.463034) |
| OmegaFold | Single-sequence prediction | Preprint evidence | [Wu et al., 2022, preprint](https://doi.org/10.1101/2022.07.21.500999) |
| Boltz-1 | Biomolecular interaction model with public artifacts | Preprint evidence | [Wohlwend et al., 2024, preprint](https://doi.org/10.1101/2024.11.19.624167) |
| Boltz-2 | Biomolecular interaction model with binding-affinity head | Preprint evidence | [Passaro et al., 2025, preprint](https://doi.org/10.1101/2025.06.14.659707) |
| Chai-1 | Biomolecular interaction model with public artifacts | Preprint evidence | [Chai Discovery et al., 2024, preprint](https://doi.org/10.1101/2024.10.10.615955) |
| AlphaFold DB | Predicted protein structure database (\>214M entries) | Demonstrated infrastructure | [Varadi et al., 2024](https://doi.org/10.1093/nar/gkad1011) |

## Protein Language Models

| Model | Main Use | Evidence Tier | Source |
|----|----|----|----|
| ESM-1b | Foundational protein language model (650M parameters, UniRef50) | Demonstrated | [Rives et al., 2021](https://doi.org/10.1073/pnas.2016239118) |
| ESM-2 | 15B-parameter protein language model | Demonstrated | [Lin et al., 2023](https://doi.org/10.1126/science.ade2574) |
| ESM3 | Multimodal generative protein language model | Demonstrated for esmGFP | [Hayes et al., 2025](https://doi.org/10.1126/science.ads0018) |
| ProtBERT / ProtTrans | BERT-style protein language model | Demonstrated baseline | [Elnaggar et al., 2022](https://doi.org/10.1109/TPAMI.2021.3095381) |
| ProGen2 | Autoregressive protein generation | Demonstrated | [Nijkamp et al., 2023](https://doi.org/10.1016/j.cels.2023.10.002) |

## Protein and Antibody Design

| Model | Main Use | Evidence Tier | Source |
|----|----|----|----|
| RFdiffusion | Protein backbone generation | Demonstrated | [Watson et al., 2023](https://doi.org/10.1038/s41586-023-06415-8) |
| RFdiffusion3 | All-atom biomolecular interaction design | Demonstrated (peer reviewed; two enzyme-design campaigns) | [Butcher et al., 2026](https://doi.org/10.1126/science.aeh2982) |
| ProteinMPNN | Inverse-folding sequence design | Demonstrated | [Dauparas et al., 2022](https://doi.org/10.1126/science.add2187) |
| LigandMPNN | Ligand-aware sequence design | Demonstrated | [Dauparas et al., 2025](https://doi.org/10.1038/s41592-025-02626-1) |
| Chroma | Programmable generative protein model | Demonstrated | [Ingraham et al., 2023](https://doi.org/10.1038/s41586-023-06728-8) |
| EvoDiff | Sequence-space diffusion | Demonstrated (preprint) | [Alamdari et al., 2023, preprint](https://doi.org/10.1101/2023.09.11.556673) |
| ProteinGenerator | Sequence-space diffusion for multistate and functional design | Demonstrated | [Lisanza et al., 2025](https://doi.org/10.1038/s41587-024-02395-w) |
| EVOLVEpro | Few-shot active learning for protein activity optimization | Demonstrated | [Jiang et al., 2025](https://doi.org/10.1126/science.adr6006) |
| COMPSS | Experiment-calibrated enzyme-design scoring | Demonstrated | [Johnson et al., 2025](https://doi.org/10.1038/s41587-024-02214-2) |
| AlphaProteo | De novo binder generation | Reported (arXiv only, restricted code) | [Zambaldi et al., 2024, preprint](https://arxiv.org/abs/2409.08022) |
| RFantibody | De novo antibody design | Demonstrated | [Bennett et al., 2026](https://doi.org/10.1038/s41586-025-09721-5) |
| IgFold | Antibody-specific structure prediction | Demonstrated | [Ruffolo et al., 2023](https://doi.org/10.1038/s41467-023-38063-x) |
| AbLang | Antibody language model | Demonstrated | [Olsen et al., 2022](https://doi.org/10.1093/bioadv/vbac046) |
| Hie PLM evolution | Protein-language-model antibody affinity maturation | Demonstrated | [Hie et al., 2024](https://doi.org/10.1038/s41587-023-01763-2) |
| Shanker structure-informed PLM | Protein and antibody complex optimization | Demonstrated | [Shanker et al., 2024](https://doi.org/10.1126/science.adk8946) |
| Rosetta3 | Classical protein design framework | Demonstrated baseline | [Leaver-Fay et al., 2011](https://doi.org/10.1016/B978-0-12-381270-4.00019-6) |

## Single-Cell and Cellular AI

| Model / Resource | Main Use | Evidence Tier | Source |
|----|----|----|----|
| scGPT | Single-cell foundation model | Demonstrated for representation; contested for perturbation | [Cui et al., 2024](https://doi.org/10.1038/s41592-024-02201-0) |
| Geneformer | Single-cell foundation model | Demonstrated | [Theodoris et al., 2023](https://doi.org/10.1038/s41586-023-06139-9) |
| scFoundation | Larger-scale scFM recipe | Demonstrated | [Hao et al., 2024](https://doi.org/10.1038/s41592-024-02305-7) |
| scBERT | Earlier scFM baseline | Demonstrated | [Yang et al., 2022](https://doi.org/10.1038/s42256-022-00534-z) |
| Nicheformer | Spatially aware single-cell and spatial-omics foundation model | Demonstrated | [Tejada-Lapuerta et al., 2025](https://doi.org/10.1038/s41592-025-02814-z) |
| Novae | Graph-based spatial transcriptomics foundation model | Demonstrated | [Blampey et al., 2025](https://doi.org/10.1038/s41592-025-02899-6) |
| SCimilarity | Cell-atlas-scale search | Demonstrated | [Heimberg et al., 2025](https://doi.org/10.1038/s41586-024-08411-y) |
| GEARS | Transcriptional perturbation prediction | Demonstrated for bounded settings | [Roohani et al., 2024](https://doi.org/10.1038/s41587-023-01905-6) |
| scGen | Perturbation transfer via VAE | Demonstrated | [Lotfollahi et al., 2019](https://doi.org/10.1038/s41592-019-0494-8) |
| CPA | Compositional perturbation autoencoder | Demonstrated | [Lotfollahi et al., 2023](https://doi.org/10.15252/msb.202211517) |
| Neural optimal transport | Counterfactual single-cell perturbation response modelling | Demonstrated | [Bunne et al., 2023](https://doi.org/10.1038/s41592-023-01969-x) |
| Perturb-CITE-seq | Joint RNA and protein perturbation screens | Demonstrated data-generation method | [Frangieh et al., 2021](https://doi.org/10.1038/s41588-021-00779-1) |
| Tabula Sapiens | Benchmark human cell atlas | Demonstrated infrastructure | [Tabula Sapiens Consortium, 2022](https://doi.org/10.1126/science.abl4896) |
| CZ CELLxGENE Census | Curated corpus of over 93 million unique cells (2025 report) | Demonstrated infrastructure | [CZI Cell Science Program et al., 2025](https://doi.org/10.1093/nar/gkae1142) |
| Cell Painting | High-content morphological profiling | Demonstrated protocol | [Bray et al., 2016](https://doi.org/10.1038/nprot.2016.105) |
| CellProfiler | Open-source image analysis | Demonstrated infrastructure | [Carpenter et al., 2006](https://doi.org/10.1186/gb-2006-7-10-r100) |
| SpatialData | Spatial omics data framework | Demonstrated infrastructure | [Marconato et al., 2025](https://doi.org/10.1038/s41592-024-02212-x) |
| JUMP morphology maps | Matched and gene-expression perturbation morphology references | Demonstrated data layer | [Chandrasekaran et al., 2024](https://doi.org/10.1038/s41592-024-02241-6); [Chandrasekaran et al., 2025](https://doi.org/10.1038/s41592-025-02753-9) |

## Histopathology, Microscopy, and Cryo-EM

| Model / Resource | Main Use | Evidence Tier | Source |
|----|----|----|----|
| UNI | Computational pathology representation learning | Demonstrated | [Chen et al., 2024](https://doi.org/10.1038/s41591-024-02857-3) |
| CONCH | Pathology visual-language representation | Demonstrated | [Lu et al., 2024](https://doi.org/10.1038/s41591-024-02856-4) |
| Virchow | Whole-slide pathology representation and rare-cancer work | Demonstrated | [Vorontsov et al., 2024](https://doi.org/10.1038/s41591-024-03141-0) |
| Prov-GigaPath | Whole-slide foundation model from real-world pathology data | Demonstrated | [Xu et al., 2024](https://doi.org/10.1038/s41586-024-07441-w) |
| CHIEF | Cancer pathology diagnosis and prognosis foundation model | Demonstrated | [Wang et al., 2024](https://doi.org/10.1038/s41586-024-07894-z) |
| MUSK | Vision-language pathology model for precision oncology | Demonstrated | [Xiang et al., 2025](https://doi.org/10.1038/s41586-024-08378-w) |
| TITAN | Multimodal whole-slide and report-aligned foundation model | Demonstrated | [Ding et al., 2025](https://doi.org/10.1038/s41591-025-03982-3) |
| HoVer-Net | Histology nuclei segmentation and classification | Demonstrated | [Graham et al., 2019](https://doi.org/10.1016/j.media.2019.101563) |
| Cellpose | Generalist cell segmentation | Demonstrated | [Stringer et al., 2021](https://doi.org/10.1038/s41592-020-01018-x) |
| Cellpose3 | Restoration-aware cell segmentation | Demonstrated | [Stringer et al., 2025](https://doi.org/10.1038/s41592-025-02595-5) |
| StarDist | Star-convex cell and nuclei detection | Demonstrated | [Schmidt et al., 2018](https://doi.org/10.1007/978-3-030-00934-2_30) |
| Segment Anything for Microscopy | Microscopy segmentation foundation model | Demonstrated | [Archit et al., 2025](https://doi.org/10.1038/s41592-024-02580-4) |
| CARE | Fluorescence microscopy restoration | Demonstrated | [Weigert et al., 2018](https://doi.org/10.1038/s41592-018-0216-7) |
| Noise2Void | Self-supervised denoising | Demonstrated | [Krull et al., 2019](https://doi.org/10.1109/CVPR.2019.00223) |
| Deep-STORM | Super-resolution localization microscopy | Demonstrated | [Nehme et al., 2018](https://doi.org/10.1364/OPTICA.5.000458) |
| ANNA-PALM | Accelerated super-resolution localization microscopy | Demonstrated | [Ouyang et al., 2018](https://doi.org/10.1038/nbt.4106) |
| In-silico labeling | Label-free fluorescent-label prediction | Demonstrated | [Christiansen et al., 2018](https://doi.org/10.1016/j.cell.2018.03.040) |
| cryoSPARC | Cryo-EM structure-determination workflow | Demonstrated | [Punjani et al., 2017](https://doi.org/10.1038/nmeth.4169) |
| Topaz | Cryo-EM particle picking | Demonstrated | [Bepler et al., 2019](https://doi.org/10.1038/s41592-019-0575-8) |
| crYOLO | Cryo-EM particle picking | Demonstrated | [Wagner et al., 2019](https://doi.org/10.1038/s42003-019-0437-z) |
| CryoDRGN | Cryo-EM heterogeneous reconstruction | Demonstrated | [Zhong et al., 2021](https://doi.org/10.1038/s41592-020-01049-4) |
| DynaMight | Cryo-EM continuous motion estimation | Demonstrated | [Schwab et al., 2024](https://doi.org/10.1038/s41592-024-02377-5) |
| tomoDRGN | Cryo-ET subtomogram heterogeneity modelling | Demonstrated | [Powell et al., 2024](https://doi.org/10.1038/s41592-024-02210-z) |

## Genome and Variant Models

| Model | Main Use | Evidence Tier | Source |
|----|----|----|----|
| Evo | 7B-parameter genome foundation model | Demonstrated | [Nguyen et al., 2024](https://doi.org/10.1126/science.ado9336) |
| Evo 2 | 40B-parameter all-domains genome model | Demonstrated | [Brixi et al., 2026](https://doi.org/10.1038/s41586-026-10176-5) |
| Nucleotide Transformer | Multi-species genome foundation model | Demonstrated | [Dalla-Torre et al., 2025](https://doi.org/10.1038/s41592-024-02523-z) |
| GET | Transcription foundation model across human cell types | Demonstrated | [Fu et al., 2025](https://doi.org/10.1038/s41586-024-08391-z) |
| Orthrus | Mature RNA foundation model | Demonstrated | [Fradkin et al., 2026](https://doi.org/10.1038/s41592-026-03064-3) |
| Semantic design with Evo | Function-guided generated genes and SynGenome | Demonstrated | [Merchant et al., 2026](https://doi.org/10.1038/s41586-025-09749-7) |
| GPN-MSA | Multispecies alignment DNA language model for variant effects | Demonstrated | [Benegas et al., 2025](https://doi.org/10.1038/s41587-024-02511-w) |
| LucaOne | Unified nucleic-acid and protein language model | Demonstrated | [He et al., 2025](https://doi.org/10.1038/s42256-025-01044-4) |
| ChatNT | Conversational agent for DNA, RNA, and protein tasks | Demonstrated | [de Almeida et al., 2025](https://doi.org/10.1038/s42256-025-01047-1) |
| AlphaGenome | Regulatory variant prediction (megabase context) | Demonstrated | [Avsec et al., 2026](https://doi.org/10.1038/s41586-025-10014-0) |
| Enformer | Long-range regulatory prediction | Demonstrated | [Avsec et al., 2021](https://doi.org/10.1038/s41592-021-01252-x) |
| DeepSEA | Single-nucleotide chromatin impact | Demonstrated | [Zhou and Troyanskaya, 2015](https://doi.org/10.1038/nmeth.3547) |
| AlphaMissense | Proteome-wide missense pathogenicity | Demonstrated as research tool | [Cheng et al., 2023](https://doi.org/10.1126/science.adg7492) |
| EVE | Evolutionary missense pathogenicity | Demonstrated | [Frazer et al., 2021](https://doi.org/10.1038/s41586-021-04043-8) |
| ESM-1v | Zero-shot variant effect from language model | Demonstrated | [Meier et al., 2021](https://proceedings.neurips.cc/paper/2021/hash/f51338d736f95dd42427296047067694-Abstract.html) |
| PrimateAI | Missense pathogenicity from primate population data | Demonstrated | [Sundaram et al., 2018](https://doi.org/10.1038/s41588-018-0167-z) |
| SpliceAI | Splice-altering variant prediction | Demonstrated | [Jaganathan et al., 2019](https://doi.org/10.1016/j.cell.2018.12.015) |
| REVEL | Missense ensemble baseline | Demonstrated baseline | [Ioannidis et al., 2016](https://doi.org/10.1016/j.ajhg.2016.08.016) |
| CADD | Cross-variant ensemble baseline | Demonstrated baseline | [Kircher et al., 2014](https://doi.org/10.1038/ng.2892) |
| Disease-specific VEP model | Cardiac and regulatory variant contexts | Demonstrated | [Zhan et al., 2025](https://doi.org/10.1038/s42256-025-01016-8) |

## Small Molecule and Drug Discovery

| Model / Resource | Main Use | Evidence Tier | Source |
|----|----|----|----|
| REINVENT 4 | Generative chemistry (RL on SMILES) | Demonstrated | [Loeffler et al., 2024](https://doi.org/10.1186/s13321-024-00812-5) |
| DiffDock | Diffusion-based blind docking | Demonstrated (with PoseBusters caveat) | [Corso et al., 2023](https://openreview.net/forum?id=kKF8_K-mBbS) |
| Pocket2Mol | Pocket-conditioned generation | Demonstrated | [Peng et al., 2022](https://proceedings.mlr.press/v162/peng22b.html) |
| Chemprop | Message-passing molecular property prediction | Demonstrated baseline | [Yang et al., 2019](https://doi.org/10.1021/acs.jcim.9b00237) |
| ADMET-AI | ADMET predictor over large libraries | Demonstrated | [Swanson et al., 2024](https://doi.org/10.1093/bioinformatics/btae416) |
| ADMETlab 3.0 | ADMET prediction platform | Demonstrated | [Fu et al., 2024](https://doi.org/10.1093/nar/gkae236) |
| Vina | Physics-based docking baseline | Demonstrated baseline | [Trott and Olson, 2010](https://doi.org/10.1002/jcc.21334) |
| Connectivity Map L1000 | Perturbational signature matching | Demonstrated infrastructure | [Subramanian et al., 2017](https://doi.org/10.1016/j.cell.2017.10.049) |
| DeepSynergy | Drug-combination synergy prediction | Demonstrated | [Preuer et al., 2018](https://doi.org/10.1093/bioinformatics/btx806) |
| SynergyFinder 2.0 | Multi-drug synergy visual analytics | Demonstrated infrastructure | [Ianevski et al., 2020](https://doi.org/10.1093/nar/gkaa216) |
| TxGNN | Graph foundation model for zero-shot drug repurposing | Demonstrated for prioritization | [Huang et al., 2024](https://doi.org/10.1038/s41591-024-03233-x) |
| Rentosertib | AI-associated TNIK inhibitor in randomized Phase 2a testing | Primary endpoint was safety and tolerability; exploratory efficacy measures | [Xu et al., 2025](https://doi.org/10.1038/s41591-025-03743-2) |

## Clinical Trial AI and Evidence Generation

| Model / Resource | Main Use | Evidence Tier | Source |
|----|----|----|----|
| TrialGPT | Trial retrieval and criterion-level eligibility assistance | Demonstrated workflow assistance on public cohorts and a small user study; not autonomous eligibility or enrollment evidence | [Jin et al., 2024](https://doi.org/10.1038/s41467-024-53081-z) |
| HINT | Retrospective phase-specific trial outcome prediction | Demonstrated retrospective research model; prospective decision impact remains unestablished | [Fu et al., 2022](https://doi.org/10.1016/j.patter.2022.100445) |
| Clinical Trial Outcome dataset | Temporally grounded trial-outcome benchmark infrastructure | Demonstrated dataset; benchmark performance does not establish better development decisions | [Gao et al., 2026](https://doi.org/10.1038/s44360-026-00081-6) |

## Autonomous Laboratories and Agentic Research Systems

| System | Main Use | Evidence Tier | Source |
|----|----|----|----|
| Coscientist | LLM-planned autonomous chemistry | Demonstrated (Pd cross-coupling) | [Boiko et al., 2023](https://doi.org/10.1038/s41586-023-06792-0) |
| Virtual Lab | Multi-agent biology research with human-executed validation | Demonstrated agentic workflow, not a robotic closed loop | [Swanson et al., 2025](https://doi.org/10.1038/s41586-025-09442-9) |
| Co-Scientist | Multi-agent biomedical hypothesis generation with selected experimental follow-up | Demonstrated in three reported applications; human experts and laboratory validation remain central | [Gottweis et al., 2026](https://doi.org/10.1038/s41586-026-10644-y) |
| Robin | Lab-in-the-loop scientific discovery system | Demonstrated in one dry age-related macular degeneration workflow; humans executed wet-lab work | [Ghareeb et al., 2026](https://doi.org/10.1038/s41586-026-10652-y) |
| A-Lab | Autonomous inorganic materials | Demonstrated platform; novelty claims debated and scoped in 2026 Author Correction (36 of 40 target syntheses confirmed; new to prediction platform) | [Szymanski et al., 2023](https://doi.org/10.1038/s41586-023-06734-w); [Leeman et al., 2024](https://doi.org/10.1103/PRXEnergy.3.011002); [Szymanski et al., 2026](https://doi.org/10.1038/s41586-025-09992-y) |
| Ada | Self-driving thin-film materials | Demonstrated | [MacLeod et al., 2020](https://doi.org/10.1126/sciadv.aaz8867) |
| ChemOS | Workflow orchestration | Demonstrated infrastructure | [Roch et al., 2020](https://doi.org/10.1371/journal.pone.0229862) |
| Mobile robotic chemist | Untethered chemistry robot | Demonstrated | [Burger et al., 2020](https://doi.org/10.1038/s41586-020-2442-2) |
| ChemCrow | LLM agent with chemistry tools | Demonstrated | [M. Bran et al., 2024](https://doi.org/10.1038/s42256-024-00832-8) |
| CellVoyager | Autonomous single-cell analysis agent | Demonstrated | [Alber et al., 2026](https://doi.org/10.1038/s41592-026-03029-6) |
| Paper2Agent | Converts a research paper plus codebase into an MCP server connected to a chat agent | Demonstrated dissemination infrastructure on AlphaGenome, Scanpy, and TISSUE; not wet-lab discovery-rate evidence | [Miao et al., 2026](https://doi.org/10.1038/s41586-026-11044-y) |
| Adam | First closed-loop robot scientist (functional genomics) | Demonstrated (historical) | [King et al., 2009](https://doi.org/10.1126/science.1165620) |
| PyLabRobot | Hardware-agnostic liquid-handling interface | Demonstrated infrastructure | [Wierenga et al., 2023](https://doi.org/10.1016/j.device.2023.100111) |
| SBOL 3 | Structured language for synthetic biology designs | Demonstrated standard | [Baig et al., 2020](https://doi.org/10.1515/jib-2020-0017) |

## Benchmarks and Evaluation

| Benchmark | Domain | Source |
|----|----|----|
| CASP | Protein structure prediction (biennial blinded) | [Kryshtafovych et al., 2024](https://pmc.ncbi.nlm.nih.gov/articles/PMC10843301/) |
| CAMEO | Continuous structure-prediction evaluation | [CAMEO, 2026](https://www.expasy.org/resources/cameo) |
| PoseBusters | Docking pose validity | [Buttenschoen et al., 2024](https://doi.org/10.1039/D3SC04185A) |
| MoleculeNet | Molecular ML benchmark (scaffold splits) | [Wu et al., 2018](https://doi.org/10.1039/C7SC02664A) |
| Therapeutics Data Commons | Multi-task therapeutic discovery | [Huang et al., 2022](https://doi.org/10.1038/s41589-022-01131-2) |
| scIB | Single-cell integration benchmark | [Luecken et al., 2022](https://doi.org/10.1038/s41592-021-01336-8) |
| OpenProblems | Single-cell multi-task benchmark | [Luecken et al., 2025](https://doi.org/10.1038/s41587-025-02694-w) |
| Virtual Cell Challenge | Perturbation-response benchmark for virtual-cell models | [Roohani et al., 2025](https://doi.org/10.1016/j.cell.2025.06.008) |
| DOME framework | Reporting checklist for supervised ML in biology | [Walsh et al., 2021](https://doi.org/10.1038/s41592-021-01205-4) |

## Critical-Evaluation References

These independent evaluations bound how much weight to give published headline numbers.

| Critique | Finding | Source |
|----|----|----|
| Ahlmann-Eltze et al. | Linear baselines compete with deep methods on single-cell perturbation prediction | [Ahlmann-Eltze et al., 2025](https://doi.org/10.1038/s41592-025-02772-6) |
| Boiarsky et al. | Independent re-evaluation of scFM finds smaller-than-reported gains | [Boiarsky et al., 2024](https://doi.org/10.1038/s42256-024-00949-w) |
| PoseBusters | Low-RMSD docking poses fail physical and chemical validity | [Buttenschoen et al., 2024](https://doi.org/10.1039/D3SC04185A) |

## Data and Infrastructure

| Resource | Main Use | Source |
|----|----|----|
| PDB | Experimental macromolecular structures | [wwPDB, 2026](https://www.wwpdb.org/?lang=en) |
| AlphaFold DB | Predicted protein structures (\>214M) | [AlphaFold Protein Structure Database, 2026](https://alphafold.ebi.ac.uk/about) |
| ESM Metagenomic Atlas | Predicted metagenomic protein structures | [ESM Metagenomic Atlas, 2026](https://esmatlas.com/about) |
| ChEMBL | Curated bioactivity database | [Zdrazil et al., 2024](https://pubmed.ncbi.nlm.nih.gov/37933841/) |
| PubChem | Chemical substance, compound, and assay database | [Kim et al., 2023](https://pubmed.ncbi.nlm.nih.gov/36305812/) |
| Open Targets | Target-disease evidence integration | [Ochoa et al., 2021](https://pmc.ncbi.nlm.nih.gov/articles/PMC7779013/) |
| Human Cell Atlas | Reference cell atlas | [Regev et al., 2017](https://doi.org/10.7554/eLife.27041) |

## Biomedical Knowledge and Literature AI

| Model / Resource | Main Use | Evidence Tier | Source |
|----|----|----|----|
| BioBERT | Biomedical text mining | Demonstrated for text-mining benchmarks | [Lee et al., 2020](https://doi.org/10.1093/bioinformatics/btz682) |
| PubMedBERT | Biomedical language representation | Demonstrated for biomedical NLP benchmarks | [Gu et al., 2022](https://doi.org/10.1145/3458754) |
| SciBERT | Scientific-text representation | Demonstrated for scientific NLP benchmarks | [Beltagy et al., 2019](https://doi.org/10.18653/v1/D19-1371) |
| SemMedDB | Literature-mined semantic predications | Demonstrated infrastructure | [Kilicoglu et al., 2012](https://doi.org/10.1093/bioinformatics/bts591) |
| Hetionet | Drug-repurposing knowledge graph | Demonstrated graph resource | [Himmelstein et al., 2017](https://doi.org/10.7554/elife.26726) |
| PrimeKG | Precision-medicine knowledge graph | Demonstrated graph resource | [Chandak et al., 2023](https://doi.org/10.1038/s41597-023-01960-3) |
| Open Targets Platform | Target-disease evidence integration | Demonstrated infrastructure | [Ochoa et al., 2021](https://doi.org/10.1093/nar/gkaa1027) |
| BioCypher | Schema-driven biomedical knowledge representation | Demonstrated infrastructure | [Lobentanzer et al., 2023](https://doi.org/10.1038/s41587-023-01848-y) |

## Programs and Regulatory References

| Program | Scope | Source |
|----|----|----|
| NIH Bridge2AI | AI-ready biomedical data and workforce | [NIH Common Fund Bridge2AI, 2026](https://commonfund.nih.gov/bridge2ai) |
| ARPA-H IGoR | AI-supported biomedical research infrastructure | [ARPA-H IGoR, 2026](https://arpa-h.gov/explore-funding/programs/igor) |
| CZI Virtual Cells Platform | Cell foundation model infrastructure | [CZI, 2026](https://virtualcellmodels.cziscience.com/) |
| FDA AI/ML in drug development | Regulatory framing for AI in medicines | [FDA, 2026](https://www.fda.gov/science-research/science-and-research-special-topics/artificial-intelligence-and-machine-learning-aiml-drug-development) |
| FDA Real-World Evidence Program | Regulatory framing for RWE use in drug development | [FDA, 2026](https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence) |
| FDA Biomarker Qualification Program | Biomarker context-of-use qualification | [FDA, 2026](https://www.fda.gov/drugs/biomarker-qualification-program/qualifying-biomarker-through-biomarker-qualification-program) |
| FDA PAT framework | Pharmaceutical manufacturing process measurement and control | [FDA, 2004](https://www.fda.gov/regulatory-information/search-fda-guidance-documents/pat-framework-innovative-pharmaceutical-development-manufacturing-and-quality-assurance) |
| EMA reflection paper on AI | EU regulatory framing for AI in medicinal lifecycle | [EMA, 2024](https://www.ema.europa.eu/en/use-artificial-intelligence-ai-medicinal-product-lifecycle) |

## Biomanufacturing References

| Resource | Main Use | Source |
|----|----|----|
| Machine learning in bioprocess development | Review of ML applications in strain engineering, process development, scale-up, monitoring, and control | [Helleckes et al., 2023](https://doi.org/10.1016/j.tibtech.2022.10.010) |
| ML-based bioprocess optimisation review | Review of optimisation, monitoring, and control systems | [Mondal et al., 2023](https://doi.org/10.1016/j.biortech.2022.128523) |
| Digital twins in biopharma manufacturing | Digital-twin literature review | [Chen et al., 2020](https://doi.org/10.3390/pr8091088) |
| Ginkgo Bioworks | Biofoundry and autonomous-lab company context | [Ginkgo Bioworks, 2026](https://www.ginkgo.bio/) |
| Cytiva | Bioprocess equipment and life-sciences infrastructure context | [Cytiva, 2026](https://www.cytivalifesciences.com/en/us/) |
| Sartorius | Bioprocess and laboratory infrastructure context | [Sartorius, 2026](https://www.sartorius.com/en) |
| Resilience | Biomanufacturing CDMO context | [Resilience, 2026](https://resilience.com/) |

## Real-World Evidence and Biomarker Platforms

| Platform | Main Use | Source |
|----|----|----|
| Tempus | Clinical and molecular data platform | [Tempus, 2026](https://www.tempus.com/) |
| Flatiron Health | Oncology real-world data infrastructure | [Flatiron Health, 2026](https://www.flatiron.com/) |
| Komodo Health | Real-world data analytics for life sciences | [Komodo Health, 2026](https://www.komodohealth.com/) |
| Datavant/Aetion | Health data collaboration and RWE analytics context | [Datavant, 2026](https://www.datavant.com/) |

## Related Resources

- [References](../references.html): Master citation list with topic organization.
- [Glossary](../appendices/glossary.html): Definitions of terminology.
- [Quick Reference: All Chapter Summaries](../appendices/tldr-compilation.html): TL;DR callouts from every chapter.

![](../images/bryan-tegomoh-author.jpg)

[Bryan Tegomoh, MD, MPH](https://bryantegomoh.com/)

Physician-scientist and epidemiologist in health AI and biosecurity.

[Follow @BryanTegomoh](https://x.com/intent/follow?screen_name=BryanTegomoh) [Follow on LinkedIn](https://www.linkedin.com/comm/mynetwork/discovery-see-all?usecase=PEOPLE_FOLLOWS&followMember=bryan-tegomoh-md-mph) [Support](../support.html)
