Generative Access to Make-on-Demand Chemical Space Enables Ultra-Large Virtual Screening
- 1Westlake University
Abstract
Ultra-large-scale virtual screening (ULVS) increasingly relies on make-on-demand chemical spaces, but efficient access to these spaces remains challenging because their scale far exceeds what can be exhaustively enumerated and docked. Here, we present REAL-SWIT, a generative ULVS framework that learns Enamine REAL Space as a synthesizable molecular distribution and couples this generator with a target-specific scoring model to enable docking-guided exploration of REAL Space at the level of complete molecule. The generative model learned transferable features of the REAL Space distribution: approximately 96% of generated molecules were found in REAL Space, and a subset of generated molecules absent from the training release appeared in later expanded releases. In computational benchmarks, REAL-SWIT identified more molecules with favorable docking scores than representative fragment-based search strategies, including cooperative building-block combinations that fragment-level prioritization tended to miss. Experimental validation for ROCK1 yielded six biochemical inhibitors among 23 synthesized compounds, including RX-3 with an IC50 of 0.17 M. These results establish generative access to make-on-demand chemical space as a practical strategy for ULVS.
Related
- Speciesformer learns conserved cellular states for cross-species generative virtual cell modeling
- Virtual experiments bridge sequence and microscopy with generative models
- An open-access neuroimaging-compatible virtual town task for assessing human spatial navigation
- Identification of a novel NR2B-selective NMDA receptor antagonist using a virtual screening approach
- EnzyREx: deep learning enables reaction-based comprehensive enzyme screening
BibTeX
@misc{zhang2026generative,
title = {{Generative Access to Make-on-Demand Chemical Space Enables Ultra-Large Virtual Screening}},
author = {K. Zhang and Y. Sun and X. Li and Y. Wang and C. Peng and X. Jin and Q. Hu and J. Huang},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.17.752345},
url = {https://www.biorxiv.org/content/10.64898/2026.09.17.752345v1},
note = {CC BY-ND},
}
RIS
TY - UNPB
TI - Generative Access to Make-on-Demand Chemical Space Enables Ultra-Large Virtual Screening
AU - K. Zhang
AU - Y. Sun
AU - X. Li
AU - Y. Wang
AU - C. Peng
AU - X. Jin
AU - Q. Hu
AU - J. Huang
PY - 2026
DA - 2026/09/17
PB - bioRxiv
DO - 10.64898/2026.09.17.752345
UR - https://www.biorxiv.org/content/10.64898/2026.09.17.752345v1
AB - Ultra-large-scale virtual screening (ULVS) increasingly relies on make-on-demand chemical spaces, but efficient access to these spaces remains challenging because their scale far exceeds what can be exhaustively enumerated and docked. Here, we present REAL-SWIT, a generative ULVS framework that learns Enamine REAL Space as a synthesizable molecular distribution and couples this generator with a target-specific scoring model to enable docking-guided exploration of REAL Space at the level of complete molecule. The generative model learned transferable features of the REAL Space distribution: approximately 96% of generated molecules were found in REAL Space, and a subset of generated molecules absent from the training release appeared in later expanded releases. In computational benchmarks, REAL-SWIT identified more molecules with favorable docking scores than representative fragment-based search strategies, including cooperative building-block combinations that fragment-level prioritization tended to miss. Experimental validation for ROCK1 yielded six biochemical inhibitors among 23 synthesized compounds, including RX-3 with an IC50 of 0.17 M. These results establish generative access to make-on-demand chemical space as a practical strategy for ULVS.
ER -