ToxCML: A Hybrid mfCoQ-RASAR-Based Platform Integrating Consensus QSAR and Read-Across for Comprehensive Multi-Endpoint Toxicity Assessment

Authors
Fauzan Syarif Nursyafi¹, Muhammad Adnan Pramudito², Yunendah Nur Fuadah³, Rahmafatin Nurul Izza², Abdul Latif Fauzan⁵, and Ki Moo Lim¹˒⁴˒⁵*

¹ Computational Medicine Lab, Department of Medical IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, 39177, Republic of Korea
² Computational Medicine Lab, Department of IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, 39177, Republic of Korea
³ Telecommunication Engineering Study Program, School of Electrical Engineering, Telkom University Main Campus, Bandung, Indonesia
⁴ Computational Medicine Lab, Department of Biomedical Engineering, Kumoh National Institute of Technology, Gumi, 39177, Republic of Korea
⁵ Meta Heart Co., Ltd, Gumi, 39253, Republic of Korea

*Corresponding author: kmlim@kumoh.ac.kr

ToxCML is a large-scale hybrid mfCoQ‑RASAR (multi-feature Consensus quantitative Read-Across Structure–Activity Relationship) platform that explicitly integrates consensus QSAR and consensus read-across into a weight-optimized hybrid predictor for multi-endpoint toxicity assessment. The framework is designed to provide chemically contextualized, applicability-domain–aware predictions to support large-scale toxicity screening, hazard prioritization, and reduction of animal testing.