Drugon: Phase-Encoded Morgan Fingerprints Match Pretrained Chemical Transformers in Low-Data Affinity Prediction
Abstract
Pretrained chemical language models such as ChemBERTa and MoLFormer have become the default choice for molecular property prediction. However, they require large training sets to fine-tune effectively, and their computational footprint is substantial. We present Drugon, a compact alternative based on a Phase-Encoded Morgan fingerprint combined with a LightGBM regressor. Drugon requires approximately 15,000× fewer parameters and 40× less training time than fine-tuned transformers, while achieving higher predictive accuracy on three independent ChEMBL targets (n = 1378 to 2000). On the c-Abl dataset (CHEMBL4550), Drugon reaches a Pearson r = 0.749 under 5-fold cross-validation, compared with 0.605 for fine-tuned ChemBERTa and 0.551 for MoLFormer. Across all three targets, phase encoding provides a consistent, non-negative improvement over the classical binary Morgan fingerprint (∆r ∈ [0.000, 0.026]). Our results suggest that, in the low-data regime (n < 2000), fingerprint-based methods remain the most efficient and interpretable
Related
- Site-Resolved Prediction of Mayr Nucleophilicity with Reactivity-Pretrained Graph Neural Networks
- Physics-based molecular fingerprints from spectral graph theory provide efficient geometry-aware measures of chemical similarity.
- Accelerating Proton Affinity Prediction with Multi-Fidelity Machine Learning
- PCIM-DTA: pairwise conditional interaction modeling for drug–target affinity prediction under cold-start scenarios
- FLOWR.ROOT – A flow matching-based foundation model for joint multi-purpose structure-aware 3D ligand generation and affinity prediction
BibTeX
@misc{azizi2026drugon,
title = {{Drugon: Phase-Encoded Morgan Fingerprints Match Pretrained Chemical Transformers in Low-Data Affinity Prediction}},
author = {Masoud Azizi},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15009827/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009827/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Drugon: Phase-Encoded Morgan Fingerprints Match Pretrained Chemical Transformers in Low-Data Affinity Prediction
AU - Masoud Azizi
PY - 2026
DA - 2026/10/04
PB - ChemRxiv
DO - 10.26434/chemrxiv.15009827/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009827/v1
AB - Pretrained chemical language models such as ChemBERTa and MoLFormer have become the default choice for molecular property prediction. However, they require large training sets to fine-tune effectively, and their computational footprint is substantial. We present Drugon, a compact alternative based on a Phase-Encoded Morgan fingerprint combined with a LightGBM regressor. Drugon requires approximately 15,000× fewer parameters and 40× less training time than fine-tuned transformers, while achieving higher predictive accuracy on three independent ChEMBL targets (n = 1378 to 2000). On the c-Abl dataset (CHEMBL4550), Drugon reaches a Pearson r = 0.749 under 5-fold cross-validation, compared with 0.605 for fine-tuned ChemBERTa and 0.551 for MoLFormer. Across all three targets, phase encoding provides a consistent, non-negative improvement over the classical binary Morgan fingerprint (∆r ∈ [0.000, 0.026]). Our results suggest that, in the low-data regime (n < 2000), fingerprint-based methods remain the most efficient and interpretable
ER -