Similarity Metrics for Ligands Reflecting the Similarity of the Target Proteins

Abstract

In this study we evaluate how far the scope of similarity searching can be extended to identify not only ligands binding to the same target as the reference ligand(s) but also ligands of other homologous targets without initially known ligands. This "homology-based similarity searching" requires molecular representations reflecting the ability of a molecule to interact with target proteins. The Similog keys, which are introduced here as a new molecular representation, were designed to fulfill such requirements. They are based only on the molecular constitution and are counts of atom triplets. Each triplet is characterized by the graph distances and the types of its atoms. The atom-typing scheme classifies each atom by its function as H-bond donor or acceptor and by its electronegativity and bulkiness. In this study the Similog keys are investigated in retrospective in silico screening experiments and compared with other conformation independent molecular representations. Studied were molecules of the MDDR database for which the activity data was augmented by standardized target classification information from public protein classification databases. The MDDR molecule set was split randomly into two halves. The first half formed the candidate set. Ligands of four targets (dopamine D2 receptor, opioid delta-receptor, factor Xa serine protease, and progesterone receptor) were taken from the second half to form the respective reference sets. Different similarity calculation methods are used to rank the molecules of the candidate set by their similarity to each of the four reference sets. The accumulated counts of molecules binding to the reference target and groups of targets with decreasing homology to it were examined as a function of the similarity rank for each reference set and similarity method. In summary, similarity searching based on Unity 2D-fingerprints or Similog keys are found to be equally effective in the identification of molecules binding to the same target as the reference set. However, the application of the Similog keys is more effective in comparison with the other investigated methods in the identification of ligands binding to any target belonging to the same family as the reference target. We attribute this superiority to the fact that the Similog keys provide a generalization of the chemical elements and that the keys are counted instead of merely noting their presence or absence in a binary form. The second most effective molecular representation are the occurrence counts of the public ISIS key fragments, which like the Similog method, incorporates key counting as well as a generalization of the chemical elements. The results obtained suggest that ligands for a new target can be identified by the following three-step procedure: 1. Select at least one target with known ligands which is homologous to the new target. 2. Combine the known ligands of the selected target(s) to a reference set. 3. Search candidate ligands for the new targets by their similarity to the reference set using the Similo...

Similarity-based virtual screening using 2D fingerprints

“…There is, however, an alternative approach to data fusion that was first suggested by Xue et al [38] and by Schuffenhauer et al [68]. This approach, which we refer to as group fusion, can be used when several, structurally-diverse reference structures are available, as may be the case from analysis of published competitor compounds or from the hits in an HTS experiment.…”

Support-Vector-Machine-Based Ranking Significantly Improves the Effectiveness of Similarity Searching Using 2D Fingerprints and Multiple Reference Compounds

“…However, search performance usually improves when multiple active compounds are available. Accordingly, various approaches have been introduced to utilize multiple reference molecules in fingerprint calculations, including consensus 9 or centroid 10 fingerprints, scaling procedures, 11 and nearest-neighbor methods 10,12 (i.e., data fusion techniques). Several studies have been conducted to compare these different search strategies, and nearestneighbor as well as centroid calculations were often found to perform best.…”

“…Several studies have been conducted to compare these different search strategies, and nearestneighbor as well as centroid calculations were often found to perform best. 10,12 The relative performance of these search strategies is often influenced by differences in structural diversity between compound classes. For example, for structurally homogeneous classes, the 1-NN approach easily detects active compounds, whereas averaging of fingerprints or similarity values often produces better results than 1-NN for moderately diverse classes.…”

Similarity‐based data mining in files of two‐dimensional chemical structures using fingerprint measures of molecular resemblance

“…The first detailed study of this type was by Willett and Winterman [12], who found that computed molecular similarities could be used to predict a range of physical, chemical and biological properties in a range of small datasets for which both structural and property information were available. There have been many subsequent examples of this approach to the evaluation of similarity procedures [13][14][15][16][17], and further supporting evidence for the general applicability of the Principle comes from studies in chemogenomics [18][19][20][21]. It must be emphasized that there are many exceptions to the Principle [22,23], but it has been found to provide a very useful basis for the development of a range of similarity-based approaches for the processing of large chemical databases.…”