LIT-PCBA
Virtual-screening benchmark whose actives and inactives are both experimentally measured, taken from high-confidence PubChem BioAssay dose-response data.
Results are recorded on the linked protocols, tasks or configurations listed below. Their settings and evaluations are kept separate.
Overview
An explanatory profile has not yet been reviewed for this record.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Results on linked records
This benchmark has no results of its own in this release. One linked record has results. They are not assigned to the underlying benchmark: each record keeps its own settings and evaluations, and its scores are on its own page.
- LIT-PCBA five-target screening (Shen et al. 2025 Table 4) Protocol · part of this benchmark · 7 evaluations · 84 results
Methods and evaluation design
Procedure, tasks and evaluated configurations
Baseline coverage
Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.
0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.
Baseline status by linked protocol
- LIT-PCBA five-target screening (Shen et al. 2025 Table 4) · 0/2 roles measured
Protocol coverage CSV (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)
Coverage is derived from release 2026-10-22-1b7cf027e4f6. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.
Run instructions
No runnable recipe has been reviewed for this benchmark. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
Strengths, limitations and unresolved questions
Evidence
Source checking verifies the cited claim or transcription. It does not establish 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.
0 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-10-22-1b7cf027e4f6 · Record review: source checked
3 source records and release history
- LIT-PCBA official project page, drugdesign.unistra.fr · Original source · Page as retrieved 2026-10-22; no version is printed on the page
- Shen et al. 2025, Chemical Science (Europe PMC full text) · Original source · PMC12704515 (Chem Sci, published 2025-12-11)
- Huang, Knight and Naprienko 2025, arXiv:2507.21404v2 (abstract page) · Original source · arXiv:2507.21404v2, submitted 2025-07-29, this version 2025-08-07
Technical metadata and extraction receipts
Stable ID: hitdisc-20261022-benchmark-litpcba
- areas
- molecular-interactions
- contexts
- research
- entity level
- suite
- task
- Protein-ligand virtual screening (library ranking)
- version
- Full data and AVE-unbiased data as published on the official page; no version string is printed
- scope note
- 15 target sets. The official page states 7761 actives and 382674 unique inactives from high-confidence PubChem BioAssay data, with per-target counts and an AVE-unbiased split published alongside the full data; Shen et al. 2025 state 10030 confirmed actives and 2798737 confirmed inactives for the full set. Because the inactives are measured rather than synthetic, and because the actives-to-inactives ratio reaches about 1:18000 for ADRB2, enrichment factors on LIT-PCBA sit far below those on property-matched decoy benchmarks.
- source locator
- Official page data content table; Shen et al. 2025 Methods 'Datasets'
- limitations
- Huang, Knight and Naprienko 2025 audit the benchmark and report extensive leakage between its splits: 323 active training-validation analogue pairs at ECFP4 Tanimoto at or above 0.6 in ALDH1 alone, and 2491 2D-identical inactives shared between training and validation across targets. They conclude that a parameter-free memorisation baseline matches or exceeds published deep-learning and 3D-similarity results, that nearly all published LIT-PCBA results are undermined, that zero-shot evaluations are also affected by analogue leakage into the query set, and that the benchmark in its current form does not measure recovery of novel chemotypes.; The official page and Shen et al. 2025 give different active and inactive totals and neither reconciles them.; The LIT-PCBA primary paper (doi 10.1021/acs.jcim.0c00155) is closed access and was not retrieved.
- review
- method: scope-review; automated-source-review; reviewer: claude; reviewer note: Claude review agent, independent of the extraction agent; no human review claimed; date: 2026-10-22; method note: Read the record against the pinned sources and against docs/use-cases.md and docs/record-contract.md.; note: Record checked against the pinned source and the record contract. Transcription checked; not reproduced and not reviewed by a human scientist.; artifact sha256: a2203612b553236d7f01da5573389d439d2f7aa0436080412f27da882e086e5e; retrieval url: https://drugdesign.unistra.fr/LIT-PCBA/; receipt: data/omics/use-case-coverage-hitdisc-20261022/review.json