Similarity-based virtual screening using 2D fingerprints

Abstract

Teaser This paper discusses the use of binary-encoded fragment substructures to scan databases to find molecules that are structurally similar to a bioactive query compound.Abstract This paper summarises recent work at the University of Sheffield on virtual screening methods that use 2D fingerprint measures of structural similarity. A detailed comparison of a large number of similarity coefficients demonstrates that the well-known Tanimoto coefficient remains the method of choice for the computation of fingerprintbased similarity, despite possessing some inherent biases related to the sizes of the molecules that are being sought. Group fusion involves combining the results of similarity searches based on multiple reference structures and a single similarity measure. We demonstrate the effectiveness of this approach to screening, and also describe an approximate form of group fusion, turbo similarity searching, that can be used when just a single reference structure is available.

Chemoinformatics at the University of Sheffield 2002–2014

“…Similarity searching is one of the most popular forms of ligand-based virtual screening [13][14][15][16] and has been one of our principal research areas for many years. Indeed, the widespread use of 2D fingerprints and the Tanimoto coefficient for computing molecular similarity is arguably due in large part to one of the first operational systems for similarity searching that was developed in a Sheffield collaboration with Pfizer in the mid-Eighties.…”

“…[13] However, it was clear in both cases that the effectiveness of this particular coefficient for virtual screening could, in some circumstances, be strongly affected by the precise nature of the weighting scheme that was being employed. This characteristic of the coefficient was investigated in a subsequent study by Holliday et al who showed that a related similarity coefficient, the cosine coefficient, was less dependent on the precise nature of the weighing scheme that was being employed and that it was, accordingly, to be preferred if weighted fingerprints were to be used for similarity searching.…”

Molecular Graph Augmentation with Rings and Functional Groups

“…The Tanimoto kernel (considered state-of-the-art for the classification of small molecules 25 ) computes a similarity score by counting the number of common elements (i.e. the set-intersection) between the two instances as a fraction of the total number of elements that occur in both instances (i.e.…”

QSAR Classification Model for Antibacterial Compounds and Its Use in Virtual Screening

“…According to World Health Organization (WHO), the top five infectious diseases with highest death rates are lower respiratory tract infections (3.9 million), diarrhea (1.8 million), tuberculosis (1.6 million), pertussis (290 000), and tetanus (210 000). 1 Considering this staggering number of deaths, there should be a lucrative market for drug therapies for these diseases. Indeed, this was true up until the early 1990s when around 20 pharmaceutical companies were involved in antibacterial research.…”