TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB
Distribution: TDC.BBB. Scored with AUROC on TDC.BBB. Scaffold split, as the ADMET benchmark group defines it. Values are the mean and standard deviation over repeated runs.
Overview
Distribution: TDC.BBB. Scored with AUROC on TDC.BBB. Scaffold split, as the ADMET benchmark group defines it. Values are the mean and standard deviation over repeated runs.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
Benchmarks
These source-backed links do not make different protocols or scores interchangeable.
Recorded evaluations
Each evaluation records what was tested and under which conditions.
Run instructions
No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.
Published comparisons
Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies. The pooled view gathers every source table that reports the same metric and names what it does not hold constant.
TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB
auroc (fraction) · Higher values are better for this metric.
Every method TDC ADMET benchmark group reports on Distribution: TDC.BBB, scored with AUROC on TDC.BBB.
Evaluation protocol · TDC.BBB (TDC ADMET benchmark group split)
- RDKit2D + MLP · Method · Author-reported evaluation0.889 ± 0.016
- Morgan + MLP · Method · Author-reported evaluation0.823 ± 0.015
- CNN · Method · Author-reported evaluation0.781 ± 0.030
Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.
Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 3, row(TDC.BBB)Values, uncertainty and evidence
| Tested entity | Printed value | Uncertainty | Evidence |
|---|---|---|---|
| RDKit2D + MLP · Method | 0.889 ± 0.016 fraction | type: standard_deviation; value: 0.016 | Author-reported evaluation · source checkedTherapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 4, block(Distribution), row(RDKit2D + MLP), column(TDC.BBB) |
| Morgan + MLP · Method | 0.823 ± 0.015 fraction | type: standard_deviation; value: 0.015 | Author-reported evaluation · source checkedTherapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 4, block(Distribution), row(Morgan + MLP), column(TDC.BBB) |
| CNN · Method | 0.781 ± 0.030 fraction | type: standard_deviation; value: 0.030 | Author-reported evaluation · source checkedTherapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 4, block(Distribution), row(CNN), column(TDC.BBB) |
Scope and limitations
- Author-reported numbers, source checked but not independently reproduced.
- MAE datasets are better when lower. The metric comes from Table 3 and differs by dataset.
- These are the paper's own simple baselines, not the current leaderboard for the ADMET group.
Source transcription and grouping reviewed by automated source review on 2026-09-18. These experiments were not independently reproduced by rewire.
Tested entities and results
Release 2026-09-17-134cd1815de8 · 3 evaluations · 3 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| CNN on TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB Method: CNNTask: TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBBDataset subset: TDC.BBB (TDC ADMET benchmark group split) Scaffold split, as the ADMET benchmark group defines it. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.781 ± 0.030 auroc Unit: fraction · Direction: higher | Uncertainty: type: standard deviation; value: 0.030 Scored: Not reported · Eligible: Not reported | source checkedTherapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 4, block(Distribution), row(CNN), column(TDC.BBB) Source checking is not independent reproduction. |
| Morgan + MLP on TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB Method: Morgan + MLPTask: TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBBDataset subset: TDC.BBB (TDC ADMET benchmark group split) Scaffold split, as the ADMET benchmark group defines it. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.823 ± 0.015 auroc Unit: fraction · Direction: higher | Uncertainty: type: standard deviation; value: 0.015 Scored: Not reported · Eligible: Not reported | source checkedTherapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 4, block(Distribution), row(Morgan + MLP), column(TDC.BBB) Source checking is not independent reproduction. |
| RDKit2D + MLP on TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB Method: RDKit2D + MLPTask: TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBBDataset subset: TDC.BBB (TDC ADMET benchmark group split) Scaffold split, as the ADMET benchmark group defines it. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.889 ± 0.016 auroc Unit: fraction · Direction: higher | Uncertainty: type: standard deviation; value: 0.016 Scored: Not reported · Eligible: Not reported | source checkedTherapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Table 4, block(Distribution), row(RDKit2D + MLP), column(TDC.BBB) Source checking is not independent reproduction. |
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: part of discovery-benchmark-tdc-molecular-tasks Individual claims | Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development Table 3, row(TDC.BBB) Version: arXiv 2102.09548v1; later NSF v2 discovery retrieved partially but full artifact unavailable | source checked automated source review · 2026-09-18 Audit detailsPrimary-source transcription with no human sign-off and no independent reproduction. Field: Claim: tdc-association-tdc-bbb Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
Release 2026-09-17-134cd1815de8 · Record review: source checked
1 source records and release history
- Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development · Original source · arXiv 2102.09548v1; later NSF v2 discovery retrieved partially but full artifact unavailable
Technical metadata and extraction receipts
Stable ID: tdc-task-tdc-bbb
- areas
- molecular-interactions
- tasks
- Distribution: TDC.BBB
- metric
- AUROC
- metric direction
- higher
- dataset
- TDC.BBB
- protocol
- Scaffold split, as the ADMET benchmark group defines it. Values are the mean and standard deviation over repeated runs.
- source locator
- Table 3, row(TDC.BBB)
- comparison panels
- id: tdc-panel-tdc-bbb; title: TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB; protocol id: tdc-task-tdc-bbb; dataset id: tdc-dataset-tdc-bbb; metric: auroc; unit: fraction; direction: higher; result ids: tdc-result-rdkit2d-plus-mlp-tdc-bbb-auroc; tdc-result-morgan-plus-mlp-tdc-bbb-auroc; tdc-result-cnn-tdc-bbb-auroc; source ids: evidence-expansion-tdc-cached-v1-aaa5526f; source locator: Table 3, row(TDC.BBB); context: Every method TDC ADMET benchmark group reports on Distribution: TDC.BBB, scored with AUROC on TDC.BBB.; caveats: Author-reported numbers, source checked but not independently reproduced.; MAE datasets are better when lower. The metric comes from Table 3 and differs by dataset.; These are the paper's own simple baselines, not the current leaderboard for the ADMET group.; review: method: automated_source_review; date: 2026-09-18
Related records
- part of: TDC molecular tasks
- subject: TDC ADMET benchmark group TDC-BBB: part of discovery-benchmark-tdc-molecular-tasks
- benchmark: CNN on TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB
- benchmark: Morgan + MLP on TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB
- benchmark: RDKit2D + MLP on TDC ADMET benchmark group TDC-BBB: Distribution: TDC.BBB