Condensing Molecular Docking CNNs via Knowledge Distillation

29 March 2024, Version 1
This content is a preprint and has not undergone peer review at the time of posting.


Knowledge of the bound protein-ligand structure is critical to many drug discovery tasks. One tool for in silico bound structure elucidation is molecular docking, which samples and scores ligand binding conformations. Recent work has demonstrated that convolutional neural networks (CNNs) for protein-ligand pose scoring outperform conventional scoring functions. Scoring performance can be further increased by taking the average of multiple CNN models, termed ensembles. However, ensembles of large parameter models require significant computational resources and therefore are difficult to apply to high-throughput molecular docking for virtual screening. We investigate knowledge distillation as a framework to condense the knowledge of large, powerful CNN model ensembles into a single reduced CNN model for a significant reduction in computational cost. Ensemble KD produces single models that outperform non-KD trained single models.


Molecular Docking
Structure-based Drug Design
Deep Learning

Supplementary materials

Supporting Information
Additional training details, extended information about evaluation metrics, list of PDB IDs for benchmarking molecular docking, additional Figures and Tables.

Supplementary weblinks


Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.