Ranking Enzyme Structures in the PDB by Bound Ligand Similarity to Biological Substrates

Abstract

SummaryThere are numerous applications that use the structures of protein-ligand complexes from the PDB, such as 3D pharmacophore identification, virtual screening, and fragment-based drug design. The structures underlying these applications are potentially much more informative if they contain biologically relevant bound ligands, with high similarity to the cognate ligands. We present a study of ligand-enzyme complexes that compares the similarity of bound and cognate ligands, enabling the best matches to be identified. We calculate the molecular similarity scores using a method called PARITY (proportion of atoms residing in identical topology), which can conveniently be combined to give a similarity score for all cognate reactants or products in the reaction. Thus, we generate a rank-ordered list of related PDB structures, according to the biological similarity of the ligands bound in the structures.

Selection of protein conformations for structure-based polypharmacology studies

“…Interestingly, these considerations are, to some extent, in line with those reported in recent papers by Tyzack J.D. et al and Das S. et al, in which the authors compared ligand-enzyme complexes from the PDB on the basis of the similarity of the bound ligands and discussed on how such results could be used to select the most relevant complexes for structure-based drug design [41,42]. Nevertheless, the selection of the most suitable protein conformations on the basis of the similarity of the co-crystallized ligands obviously presents some limitations.…”

Finding enzyme cofactors in Protein Data Bank

“…Small molecules that are structurally similar to the cofactor template molecules (Supplementary Table S3) are identified. A chemical structure similarity score is calculated using RDKit-based similarity-searching methods PARITY (Tyzack et al ., 2018) and SiteBinder (Sehnal et al ., 2012). Small molecules with a similarity score above a predefined cut-off specified for a particular cofactor class are tentatively selected for further manual inspection for appropriate structural equivalence with the template molecule (Supplementary Table S3) and are added to the list of small molecules in the corresponding cofactor classes.An automatic process obtains a list of PDB entries containing newly identified cofactor-like small molecule and enzymes associated with the corresponding cofactor classes.…”

“…Small molecules that are structurally similar to the cofactor template molecules (Supplementary Table S3) are identified. A chemical structure similarity score is calculated using RDKit-based similarity-searching methods PARITY (Tyzack et al ., 2018) and SiteBinder (Sehnal et al ., 2012).…”

“…Relying on the knowledge of reactions catalysed by these enzymes, the classification of cofactors into 27 classes in the CoFactor database (Fischer et al. , 2010) and a new method to measure molecular similarity (Tyzack et al ., 2018), we have developed a protocol to identify cofactors and cofactor-like molecules in cofactor-dependent enzymes.…”

CATH functional families predict protein functional sites

“…cognate and non-cognate) and (ii) differences in the characteristics of ligand binding between enzymes and nonenzymatic proteins. For example, it is known that non-cognate ligands can bind to different regions in a protein to those bound by the cognate ligands, reflecting different binding characteristics (Tyzack, Fernando, Ribeiro, Borkakoti, & Thornton, 2018). Figure 3d shows the performance of the LIG predictors on the LIG hold-out test set compared to the ConCavity ligand-binding site prediction method.…”