AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning
Zhikang Yuan, Xin Zhang, Gaoming Lin, Quan Zou, Subhashisa Swain, Yijie Ding, Prayag Tiwari, Shuofeng Yuan, Xiaoyi Guo · 2026
Read AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning by Zhikang Yuan, Xin Zhang, Gaoming Lin, Quan Zou, Subhashisa Swain, Yijie Ding, Prayag Tiwari, Shuofeng Yuan, Xiaoyi Guo (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-13561366.txt.
Description
Zhikang Yuan and co-authors design in a 2026 study new drug candidates from gene activity patterns, using the signature of a disease being reversed as the blueprint. Expression data is sparse and noisy, which blurs the map from signature to molecule. Their framework, AET5, first learns noise-tough signature features through contrastive training, then aligns them with molecule sequences.
On a large public data set it beat rival expression-guided generators on quality and realism while keeping drug-like traits. Candidates for COVID-19 and prostate cancer docked stably to their targets in simulation. Reversal signatures, they show, can steer molecule invention.
Catalog record
| CREATOR | Zhikang Yuan, Xin Zhang, Gaoming Lin, Quan Zou, Subhashisa Swain, Yijie Ding, Prayag Tiwari, Shuofeng Yuan, Xiaoyi Guo |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 77 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |