Superwater as a generative AI framework to predict water molecule positions on protein structures
Xiaohan Kuang, Yunchao Lance Liu, Xiaobo Lin, Jesse Spencer-Smith, Tyler Derr, Yinghao Wu, Hans Bitter, Yongbo Hu, Jens Meiler, Zhaoqian Su · 2025
Read Superwater as a generative AI framework to predict water molecule positions on protein structures by Xiaohan Kuang, Yunchao Lance Liu, Xiaobo Lin, Jesse Spencer-Smith, Tyler Derr, Yinghao Wu, Hans Bitter, Yongbo Hu, Jens Meiler, Zhaoqian Su (2025) 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-12715242.txt.
Description
Water molecules hug every protein, steadying its shape and guiding its meetings with other molecules, yet their exact spots are hard to know without a crystal. Xiaohan Kuang and nine co-authors present a generative model that predicts those spots in this 2025 paper. The model, called SuperWater, pairs a diffusion process with a graph network that respects the symmetry of three-dimensional space.
SuperWater beat older methods on both coverage of known crystal waters and the precision of its placements, landing within a fraction of an atomic width of the measured spots on average. Case studies show it at work on plain hydration, on drug binding pockets, and on the seams where two proteins meet. The authors offer the tool for jobs from basic structure studies to water-aware drug design.
Catalog record
| CREATOR | Xiaohan Kuang, Yunchao Lance Liu, Xiaobo Lin, Jesse Spencer-Smith, Tyler Derr, Yinghao Wu, Hans Bitter, Yongbo Hu, Jens Meiler, Zhaoqian Su |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |