A generative framework for predicting cellular morphological and transcriptomic perturbation responses
Rui Peng, Ziru Liu, Yuanxu Gao, Jinzhuo Wang · 2026
Read A generative framework for predicting cellular morphological and transcriptomic perturbation responses by Rui Peng, Ziru Liu, Yuanxu Gao, Jinzhuo Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13390005.txt.
Description
Peng, Liu, Gao, and Wang published in twenty twenty-six a generative model that predicts how cells respond to drugs and gene changes. Drug hunters need to foresee changes in both cell shape and gene activity. Past methods predict one or the other, or need data taken after treatment. Their model, MultiVCDiff, predicts both at once from the treatment alone. It trained on more than a thousand chemical and genetic changes across four data sets.
Tested on unseen drugs with no prior data, it beat single-output rivals. It avoided the common collapse where generated cells all look alike. Its gene signatures were accurate enough to recover how drugs act. The authors say such virtual screening can speed drug discovery. It also advances the goal of full virtual cell models.
Catalog record
| CREATOR | Rui Peng, Ziru Liu, Yuanxu Gao, Jinzhuo Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 76 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |