{"id":"arxiv/2607.11701","url":"https://arxiv.org/abs/2607.11701","tags":["quant-ph","cs.LG"],"score":0,"title":"$\\mathtt{Q^2SAR}$: overcoming classical bottlenecks in drug discovery via quantum multiple kernel learning","domain":"quant-ph","source":"arxiv","authors":["Mariano Caruso","Daniel Ruiz","Alejandro Giraldo","Guido Bellomo"],"reviews":0,"subject":"quant-ph","summary":"Quantitative Structure-Activity Relationship ($\\mathtt{QSAR}$) modeling is a foundational computational methodology in early-stage drug discovery, heavily relied upon for predicting compound toxicity, bioavailability, and therapeutic potential. However, classical methods often struggle to effectively map the highly complex, non-linear, and high-dimensional interactions inherent in molecular data, leading to reduced predictive accuracy and costly late-stage clinical failures. In this paper, we present a Quantum Multiple Kernel Learning ($\\mathtt{QMKL}$) framework, dubbed Next-Gen…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-07-13","reactions":{}}