An Integrated Deep Learning and Molecular Dynamics Simulation-Based Screening Pipeline Identifies Inhibitors of a New Cancer Drug Target TIPE2.
Source check: Source checked. Not independent scientific confirmation.
Finding: A hybrid deep-learning, docking and molecular-dynamics screen nominated TIPE2 ligands. Biolayer interferometry tested four: UM-164 bound at Kd 4.97 μM, while three alternatives showed much weaker reported binding. A PIP2 competition assay does not establish cancer-treatment efficacy.
Technology: DFCNN dense fully connected neural network
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- DFCNN and DeepBindBC binding predictors, docking and molecular-dynamics simulations prioritized compounds for experimental TIPE2 binding and PIP2-competition assays.
- Test and result
- BLI tested four selected compounds: UM-164 Kd 4.97 μM, golvatinib about 189 μM and fedratinib about 298 μM. FIIN-3 was described as lacking reactivity; the discussion’s approximately 0.858 M fit is not evidence of appreciable binding. A separate UM-164 competition assay reduced TIPE2–PIP2 binding.
- Limitation / still unknown
- The assays establish biochemical binding and in-vitro competition only, not cellular target engagement or treatment efficacy. The paper's conclusion calls UM-164 low-nanomolar, but the reported BLI Kd is 4.97 μM.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.3389/fphar.2021.772296
Source checked · editorial details
The source page was checked against this entry. 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: Frontiers primary Methods on DFCNN/DeepBindBC and Results/Discussion on BLI binding, competition, and internal affinity inconsistencies..
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.