AI Science Tracker

China Pharmaceutical University

3 discoveries from 2020 to 2026

Discoveries

Integrating Medicinal Chemist Expertise with Deep Learning for Automated Molecular Optimization.

Source check: Needs source check. Not independent scientific confirmation.

Finding / tested result: Two optimized compounds improved biochemical activity against cancer-related targets; expert-derived strategies also contributed.

Technology: MolOpt graph deep learning; AutoOptimizer

Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
MolOpt graph deep learning; AutoOptimizer is the AI method documented in the editorial source review.
Limitation / still unknown
The date refers to the cited version. Earlier public versions and independent replication were not systematically verified.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://doi.org/10.1021/acs.jmedchem.5c03746

Needs source check · editorial details

The source has not been fully checked or contains conflicting information. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review scope and date are not recorded; source checked does not establish full-text review.

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Effective Reaction-Based De Novo Strategy for Kinase Targets: A Case Study on MERTK Inhibitors.

Source check: Needs source check. Not independent scientific confirmation.

Finding / tested result: Generated building blocks fed reaction-constrained design yielding new MERTK inhibitors with measured kinase activity.

Technology: Recurrent neural network

Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
Recurrent neural network is the AI method documented in the editorial source review.
Limitation / still unknown
The date refers to the cited version. Earlier public versions and independent replication were not systematically verified.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://doi.org/10.1021/acs.jcim.2c00068

Needs source check · editorial details

The source has not been fully checked or contains conflicting information. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review scope and date are not recorded; source checked does not establish full-text review.

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Discovery of Dual FGFR4 and EGFR Inhibitors by Machine Learning and Biological Evaluation.

Source check: Needs source check. Not independent scientific confirmation.

Finding: Four machine-learning methods were compared; the selected SVM models screened in-house compounds for dual FGFR4/EGFR inhibition. The primary abstract reports compound 1 kinase IC50 values of 86.2 and 83.9 nM, respectively. This assay result does not establish clinical efficacy.

Technology: Support-vector-machine activity models

Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
SVM, random forest, gradient-boost regression tree and XGBoost models were compared after mutual-information feature selection; the best-performing SVM models predicted activity of in-house compounds.
Test and result
The primary abstract reports four QSAR methods and a selected dual inhibitor with IC50 86.2 nM for FGFR4 and 83.9 nM for EGFR. Assay denominator, comparator, and cell-level validation details were not present in the reviewed abstract; full ACS text was inaccessible.
Limitation / still unknown
Full Methods and biological-evaluation details were not reviewed. The date is unresolved: ACS header September 14 versus publication-history September 24 and PubMed Epub September 23, 2020.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://doi.org/10.1021/acs.jcim.0c00652

Needs source check · editorial details

The source has not been fully checked or contains conflicting information. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 18 Nov 2026. Review scope: Indexed primary ACS/PubMed abstract and conflicting publication metadata; complete Methods/Results not reviewed..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.