A multi-feature integration algorithm for ligand-similarity-based virtual screening

LigMate is a multi-feature integration algorithm, which employs Hungarian-algorithm-based matching and machine-learning based non-linear combinations of various descriptors including new relevant descriptors focusing on the maximum common substructures (MCSS), the relative distance of atoms from the ligand mass center (ILDS) as well as the ring differences (RS). It also incorporates two terms proposed by Hu et al (2018): a weighted Jaccard score measuring the similarity of chemical bonds through which atoms interact with other atoms (BWJS) and a score measuring the difference in the relative masses (MS).

Responsive image