GenePT
Uses text embeddings of gene descriptions from a general LLM to represent cells, and performs surprisingly well.
GenePT (Stanford, 2023) showed that embeddings from GPT-3.5 gene summaries rival specialised models on many tasks, questioning how much single-cell pretraining adds.
Generic schematic · not to scale · placeholder for the ai computation front
Flagged finding · Neural network
How it works
GenePT aggregates LLM gene-description embeddings weighted by expression.
Strengths
- Cheap, interpretable
Limitations
- No perturbation modelling
Since
2023
Latest papers
topLatest papers · live from Europe PMC
Query for this technology: (TITLE:"GenePT" OR ABSTRACT:"GenePT") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about GenePT, not a curated reading list.