Therapeutics Data Commons
Verified by Mdexa · Checked against the source text under public rules.How we check
- Commercial use
- UnclearUnstated
- Derivatives
- UnclearUnstated
- Redistribution
- UnclearUnstated
- File on Mdexa
- NoPending
the source says licenses differ per subset and does not list them
Cite this pageMdexa, Therapeutics Data Commons, https://mdexa.com/datasets/tdc/, verified 2026-11-14.
Key facts
- Domain
- Pharmacy and drugs · Clinical medicine Modern medicine
- Content
- DatabasesA coordinated initiative and resource offering AI-solvable tasks, AI-ready datasets and curated benchmarks for therapeutic science, with tools, libraries, leaderboards and community resources.
- Size
- —
- Language / modality
- English / Text · Tabular
- Publisher
- Zitnik Lab, Harvard
- Year
- 2020
- Expert review
- —
- Verification
- Verified: license text seen
Commercial use of Therapeutics Data Commons
Commercial use: Unclear (Unstated). License status: Unstated.
Quoted statements and their sources are in Therapeutics Data Commons license evidence.
Verdicts record what sources state and are not legal advice.
Practical information
Content scope
- The text states that TDC includes 66 AI-ready datasets spread across 22 learning tasks and spanning the discovery and development of safe and effective medicines.
arxiv.org ·
To date, TDC includes 66 AI-ready datasets spread across 22 learning tasks and spanning the discovery and development of safe and effective medicines.
https://arxiv.org/abs/2102.09548 · preprint · checked 2026-11-13
- The text states that TDC covers learning tasks including target discovery, activity screening, efficacy, safety and manufacturing across biomedical products such as small molecules, antibodies and vaccines.
tdcommons.ai ·
TDC covers a wide range of learning tasks, including target discovery, activity screening, efficacy, safety, and manufacturing across biomedical products, including small molecules, antibodies, and vaccines.
https://tdcommons.ai/ · Dataset page · checked 2026-11-13
Identifiers
- dataset name within a task classThe text shows a dataset retrieved by the name HIA_Hou from the ADME therapeutic learning task under single-instance prediction.
raw.githubusercontent.com ·
For a concrete example, to obtain the HIA dataset from the ADME therapeutic learning task in the single-instance prediction problem:
https://raw.githubusercontent.com/mims-harvard/TDC/HEAD/README.md · README · checked 2026-11-13
Access
- Web interface onlygithub.com
Same terms as the card's license.
Versions
- LatestTDC is versioned as the PyTDC Python package; the latest PyPI release is 1.1.15, uploaded 2025-03-31. No separate data-release numbering was found.
pypi.org ·
"upload_time":"2025-03-31T15:20:14"
https://pypi.org/pypi/PyTDC/json · official-release-notes · checked 2026-11-13
- Update cycleThe text states that TDC is regularly updated with new datasets and tasks, such as antibody therapeutics and gene editing.
tdcommons.ai ·
TDC is regularly updated with novel datasets and tasks, such as antibody therapeutics and gene editing.
https://tdcommons.ai/overview/ · Dataset page · checked 2026-11-13
Known defects and misuse
- The author writes that TDC is "equally flawed" as MoleculeNet, judged by the criteria he sets out in the article.
practicalcheminformatics.blogspot.com ·
other widely used benchmarks like the Therapeutic Data Commons (TDC) are equally flawed.
https://practicalcheminformatics.blogspot.com/2023/08/we-need-better-benchmarks-for-machine.html · practitioner-article · checked 2026-11-13
- Walters: the MoleculeNet BBB set contains 59 duplicate structures, 10 of which carry conflicting labels, plus 11 SMILES RDKit cannot parse.
practicalcheminformatics.blogspot.com ·
First, the dataset contains 59 duplicate structures. This is not the sort of thing we want to have in a benchmark. But wait, it gets worse, 10 of these duplicate structures have different labels.
https://practicalcheminformatics.blogspot.com/2023/08/we-need-better-benchmarks-for-machine.html · practitioner-article · checked 2026-11-13
- Walters: in the MoleculeNet BACE set 71% of molecules have at least one undefined stereocenter.
practicalcheminformatics.blogspot.com ·
71% of the molecules have at least one undefined stereocenter, 222 molecules have 3 undefined stereocenters, and one molecule has a whopping 12 undefined stereocenters.
https://practicalcheminformatics.blogspot.com/2023/08/we-need-better-benchmarks-for-machine.html · practitioner-article · checked 2026-11-13
Suited for
- The text states that TDC is a resource to access and evaluate AI methods and supports the development of AI methods for drug discovery applications.
tdcommons.ai ·
Therapeutics Data Commons is a resource to access and evaluate AI methods, supporting the development of AI methods, with a strong bent towards establishing the foundation of which AI methods are most suitable for drug discovery applications and why.
https://tdcommons.ai/ · Dataset page · checked 2026-11-13
- TDC authors state their experiments show even the strongest algorithms fall short on distribution shift and generalisation to novel data.
arxiv.org ·
demonstrating that even the strongest algorithms fall short of solving key therapeutics challenges, including real dataset distributional shifts, multi-scale modeling of heterogeneous data, and robust generalization to novel data points.
https://arxiv.org/abs/2102.09548 · preprint · checked 2026-11-13
- Walters: MoleculeNet physiology datasets (BBB, toxicology) have complex endpoints that make them less than ideal for benchmark comparisons.
practicalcheminformatics.blogspot.com ·
The complexity of these endpoints makes the data less than ideal for benchmark comparisons.
https://practicalcheminformatics.blogspot.com/2023/08/we-need-better-benchmarks-for-machine.html · practitioner-article · checked 2026-11-13
Therapeutics Data Commons license evidence 5
Code repository license: MIT. This licenses the repository, not the data; the data has no separate statement.
LICENSE file · verified s08
-
Repository license filePage opened
MITThird-party text · Repository license file
"spdx_id": "MIT"
GitHub repository license found in snapshot s08 (api-license)
-
Terms of use pagePage opened
MITThird-party text · Terms of use page
The TDC codebase is licensed under the MIT license.
Statement found in snapshot s05 (readme)
-
Terms of use pagePage opened
Third-party text · Terms of use page
For individual dataset usage, please refer to the dataset license on the website.
Statement found in snapshot s05 (readme)
-
Terms of use pagePage opened
MITThird-party text · Terms of use page
Permission is hereby granted, free of charge, to any person obtaining a copy
Statement found in snapshot s06 (license)
-
Terms of use pagePage opened
MITThird-party text · Terms of use page
The TDC codebase is licensed under the MIT license. For individual dataset usage, please refer to the dataset license on the website.
Statement found in snapshot s07 (page)
Source and lineage
No upstream or downstream datasets recorded.
Known issues with Therapeutics Data Commons
- The README defers per-dataset licensing to the website, so the license of the individual datasets is not stated here.
- The release is a beta release, so the local copy should be updated regularly.
Preflight
Not run Runs only on ingested data; says nothing about quality.
Downloading Therapeutics Data Commons and API access
No file on Mdexa (see the verdict). Get it from the source and confirm the terms there:
Citing Therapeutics Data Commons
Mdexa, Therapeutics Data Commons, https://mdexa.com/datasets/tdc/, verified 2026-11-14.
More citation details
The source gives no citation text; the line above cites this card.
Machine-readable
Change log 10
| 2026-11-14 | Known defects, version notes and access statements from cited sources (papers, release notes, official pages) were added; each shows who said it, when, and what kind of source it is. | |
| 2026-11-14 | Practical fields (scope, identifiers, access methods, versions, known defects) were added by the automated pipeline from stored page snapshots; every statement carries a verbatim quote. | |
| 2026-11-14 | Known defects, version notes and access statements from cited sources (papers, release notes, official pages) were added; each shows who said it, when, and what kind of source it is. | |
| 2026-11-14 | Known defects, version notes and access statements from cited sources (papers, release notes, official pages) were added; each shows who said it, when, and what kind of source it is. | |
| 2026-11-14 | Re-adjudicated under the current rules; all machine checks still pass. | |
| 2026-11-14 | Known defects, version notes and access statements from cited sources (papers, release notes, official pages) were added; each shows who said it, when, and what kind of source it is. | |
| 2026-11-14 | Known defects, version notes and access statements from cited sources (papers, release notes, official pages) were added; each shows who said it, when, and what kind of source it is. | |
| 2026-11-14 | Re-adjudicated under the current rules; all machine checks still pass. | |
| 2026-11-14 | Known defects, version notes and access statements from cited sources (papers, release notes, official pages) were added; each shows who said it, when, and what kind of source it is. | |
| 2026-11-13 | Added via the automated pipeline and passed all checks (gates, verbatim quotes, rule engine, two independent extractions). |
Listed in
Datasets similar to Therapeutics Data Commons
TopU-LBVS: Multi-Target Benchmark for Ligand-Based Virtual ScreeningVerified by Mdexa PermissiveLink only —
License permissive, share-alike: all parts permissive; conditions are the union of the parts. Risk flag(s): derived. Link only.
Pharmacy and drugs · Databases · Derived
Shared: Intended use Evaluation · Specialty Pharmacy · Official split Train and test splits · domain Pharmacy and drugs · content Databases
BindingDBVerified by Mdexa PermissiveLink only —
License permissive, share-alike: all parts permissive; conditions are the union of the parts. Risk flag(s): derived. Link only.
Pharmacy and drugs · Databases · Derived
Shared: Intended use Evaluation · Specialty Pharmacy · domain Pharmacy and drugs · content Databases
DrugCentralVerified by Mdexa UnstatedPending —
License unstated, judged by public rules and checked by a second-pass model. The license question stays open.
Pharmacy and drugs · Clinical medicine · Standards and terminology · Databases · Real patient text
Shared: Specialty Pharmacy · domain Pharmacy and drugs, Clinical medicine · content Databases
ClinPGx/PharmGKBVerified by Mdexa RestrictedLink only —
License restricted, judged by public rules and checked by a second-pass model. Link only.
Clinical medicine · Genomics and omics · Pharmacy and drugs · Annotated corpora · Databases · Knowledge graphs · Guidelines and standards
Shared: Specialty Pharmacy · domain Pharmacy and drugs, Clinical medicine · content Databases
医学考试题库 JSON 数据集Verified by Mdexa PermissiveLink only —
License permissive: first-hand data-scope statement(s) allow commercial use and redistribution. Risk flag(s): Exam-derived. Link only.
Clinical medicine · Pharmacy and drugs · Chinese materia medica · Medical imaging · Public health · Question banks and benchmarks · Exam-derived
Shared: Intended use Evaluation · domain Pharmacy and drugs, Clinical medicine
No sourced statement yet
Breaking changes · Unsuited tasks · Commercial licensing