Geneformer identifies cardiac gene targets validated in engineered heart tissues
Source check: Source checked. Not independent scientific confirmation.
Finding: Geneformer used learned single-cell transcriptome representations to nominate genes affecting cardiac function. Laboratory tests found that loss of the predicted dosage-sensitive candidate TEAD4 reduced contractile stress in cardiac microtissues. In a separate disease-model application, targeting the Geneformer-nominated genes GSN or PLN improved contractile stress in microtissues carrying a TTN truncating mutation. These are same-study findings in engineered cell-derived tissues, not evidence of clinical benefit. The date records the journal online version; an earlier public version was not established in the bounded search.
Technology: Geneformer attention-based pretrained transformer; in silico gene perturbation
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- A pretrained transformer supplied context-dependent cell embeddings. Simulated gene perturbations prioritized candidates; researchers performed the functional tissue experiments.
- Test and result
- Geneformer-nominated cardiac genes were tested in engineered iPSC-derived microtissues. TEAD4 loss impaired contractile stress; GSN or PLN targeting improved this readout in the TTN-mutant disease model.
- Limitation / still unknown
- Clinical safety, therapeutic efficacy and generalization beyond the studied engineered tissues were not established by these experiments. Full supplements and independent replication were not reviewed.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC10949956/ · https://escholarship.org/content/qt8st2d5mx/qt8st2d5mx.pdf · https://pubmed.ncbi.nlm.nih.gov/37258680/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 21 Nov 2026. Review scope: Primary author manuscript: bounded model/in-silico-perturbation Methods, cardiac Results and figures2/6; institutional manuscript and PubMed dates/affiliations. Supplements not fully 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 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.