DCPM-ADMET: fusion of dual-component pre-trained model and molecular fingerprints to enhance drug ADMET properties prediction
Leilei Zhang, Yuchen Zeng, Yue Qi, Kaili Jiang, Xiaofei Zhou, Lu Liang, Jianping Lin · 2026
Read DCPM-ADMET: fusion of dual-component pre-trained model and molecular fingerprints to enhance drug ADMET properties prediction by Leilei Zhang, Yuchen Zeng, Yue Qi, Kaili Jiang, Xiaofei Zhou, Lu Liang, Jianping Lin (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/PMC13540921.txt.
Description
Leilei Zhang and six colleagues build, in a 2026 study, a deep learning model that predicts how candidate drugs will be absorbed, distributed, broken down, excreted, and whether they will prove toxic. Such ADMET traits decide clinical success, yet traditional models generalise poorly when data are thin.
Their model fuses a language-style network reading molecular sequences, a second network digesting global property descriptors, and fingerprint encodings of local substructures. It outperformed traditional methods and most rival pre-trained models on standard benchmarks. Fine-tuning on a self-built database of nearly half a million entries produced ninety-seven property predictors. A free online tool now offers 133 prediction endpoints.
Catalog record
| CREATOR | Leilei Zhang, Yuchen Zeng, Yue Qi, Kaili Jiang, Xiaofei Zhou, Lu Liang, Jianping Lin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34957 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |