Paper ID: 2303.10179

QUBO-inspired Molecular Fingerprint for Chemical Property Prediction

Koichiro Yawata, Yoshihiro Osakabe, Takuya Okuyama, Akinori Asahara

Molecular fingerprints are widely used for predicting chemical properties, and selecting appropriate fingerprints is important. We generate new fingerprints based on the assumption that a performance of prediction using a more effective fingerprint is better. We generate effective interaction fingerprints that are the product of multiple base fingerprints. It is difficult to evaluate all combinations of interaction fingerprints because of computational limitations. Against this problem, we transform a problem of searching more effective interaction fingerprints into a quadratic unconstrained binary optimization problem. In this study, we found effective interaction fingerprints using QM9 dataset.

Submitted: Mar 17, 2023