BioGPS Descriptors for Rational Engineering of Enzyme Promiscuity and Structure Based Bioinformatic Analysis

Abstract

A new bioinformatic methodology was developed founded on the Unsupervised Pattern Cognition Analysis of GRID-based BioGPS descriptors (Global Positioning System in Biological Space). The procedure relies entirely on three-dimensional structure analysis of enzymes and does not stem from sequence or structure alignment. The BioGPS descriptors account for chemical, geometrical and physical-chemical features of enzymes and are able to describe comprehensively the active site of enzymes in terms of “pre-organized environment” able to stabilize the transition state of a given reaction. The efficiency of this new bioinformatic strategy was demonstrated by the consistent clustering of four different Ser hydrolases classes, which are characterized by the same active site organization but able to catalyze different reactions. The method was validated by considering, as a case study, the engineering of amidase activity into the scaffold of a lipase. The BioGPS tool predicted correctly the properties of lipase variants, as demonstrated by the projection of mutants inside the BioGPS “roadmap”.

Nature Inspired Solutions for Polymers: Will Cutinase Enzymes Make Polyesters and Polyamides Greener?

“…Similarities between CaLB and Thc_cut1 are not just limited to their surface properties and dynamic behavior but are also evident by considering their active site properties. Features such as ability of establishing H bonds and hydrophobic-hydrophilic balance were recently analyzed quantitatively by means of a bioinformatics method based on BioGPS molecular descriptors (Global Positioning System in Biological Space) and multivariate statistical analysis [55,56]. The method allows for functional structural correlation by comparing three-dimensional structures rather than protein sequences, and it makes any structural alignment unnecessary.…”

Navigating within thiamine diphosphate‐dependent decarboxylases: Sequences, structures, functional positions, and binding sites

“…The magnitude of the interaction of the N1 and O probes also includes, implicitly, information about the charge contribution, as these probes already have a partially positive and negative charge, respectively. The pseudo‐MIFs points were filtered, by means of a weighted energy‐based and space‐coverage function, and then used for the generation of quadruplets obtained from all possible combinations of the four pseudo‐MIF points. Thus, each active site was described by a series of quadruplets. The active sites of the enzymes were compared on the basis of the corresponding quadruplets, using an all‐versus‐all approach. The computational comparison generated a matrix of similarity scores (the BioGPS descriptors), which were analyzed by means of a principal component analysis (PCA) …”

Comparing Drug Images and Repurposing Drugs with BioGPS and FLAPdock: The Thymidylate Synthase Case

“…The BioGPS similarity score quantifies the geometrical and chemical similarity of multiple pockets upon alignment of their corresponding MIFs, and gives valuable clues about the structural correlation of proteins, even when belonging to distant and diverse families. Only GRID MIFs and multivariate statistical analysis are used to compare and cluster protein families . No sequence‐related information, ligand similarity or side effect relationship is needed.…”

“…Comparing the MIFs means comparing the chemical and geometrical properties as well as the encoded energetics. Recent applications have demonstrated BioGPS’ capability of predicting off‐target effects, classifying protein families, justifying polypharmacology, and rationalizing selectivity between sub‐families …”

“…[44] Pockets are describedb yt heir molecular interaction fields (MIFs, calculated by GRID [53,54] ), that is, the shape, the hydrophobic regions, the H-bond donor and acceptorh otspots al igand would encounteru pon enteringt he cavity.T he BioGPS similarity score quantifies the geometrical and chemical similarityo fm ultiple pockets upon alignment of their corresponding MIFs, and gives valuable clues aboutt he structural correlation of proteins, even when belongingt od istant and diverse families.Only GRID MIFs and multivariate statistical analysis are used to compare and cluster protein families. [55] No sequence-related information, ligand similarity or side effectr elationship is needed. With respect to other methods based on pseudocenters, which only represent the hydrophobic or Hbond donor/acceptor nature of the residues lining ac avity, BioGPS considers the MIFs generated by those residues according to their environment and, thus, the energetics of ap ocket.…”