Biology subjects

Kirubakaran, P.

Publications and source records attributed to Kirubakaran, P..

3 recordsLinked to original sources

bioRxiv2023

Poor Generalization by Current Deep Learning Models for Predicting Binding Affinities of Kinase Inhibitors

The extreme surge of interest over the past decade surrounding the use of neural networks has inspired many groups to deploy them for predicting binding affinities of drug-like molecules to their receptors. A model that can accurately make such predictions has the potential to screen large chemical libraries and help streamline the drug discovery process. However, despite reports of models that accurately predict quantitative inhibition using protein kinase sequences and inhibitors SMILES strings, it is still unclear whether these models can generalize to previously unseen data. Here, we build a Convolutional Neural Network (CNN) analogous to those previously reported and evaluate the model over four datasets commonly used for inhibitor/kinase predictions. We find that the model performs comparably to those previously reported, provided that the individual data points are randomly split between the training set and the test set. However, model performance is dramatically deteriorated when all data for a given inhibitor is placed together in the same training/testing fold, implying that information leakage underlies the models performance. Through comparison to simple models in which the SMILES strings are tokenized, or in which test set predictions are simply copied from the closest training set data points, we demonstrate that there is essentially no generalization whatsoever in this model. In other words, the model has not learned anything about molecular interactions, and does not provide any benefit over much simpler and more transparent models. These observations strongly point to the need for richer structure-based encodings, to obtain useful prospective predictions of not-yet-synthesized candidate inhibitors.

bioinformatics↗

bioRxiv2020

Comparative Modeling of CDK9 Inhibitors to Explore Selectivity and Structure-Activity Relationships

Cyclin-dependent kinase 9 (CDK9) plays a key role in transcription elongation, and more recently it was also identified as the molecular target of a series of diaminothiazole compounds that reverse epigenetic silencing in a phenotypic assay. To better understand the structural basis underlying these compounds activity and selectivity, we developed a comparative modeling approach that we describe herein. Briefly, this approach draws upon the strong structural conservation across the active conformation of all protein kinases, and their shared pattern of interactions with Type I inhibitors. Because of this, we hypothesized that the large collection of inhibitor/kinase structures available in the Protein Data Bank (PDB) would enable accurate modeling of this diaminothiazole series in complex with each CDK family member. We apply this new comparative modeling pipeline to build each of these structural models, and then demonstrate that these models provide retrospective rationale for the structure-activity relationships that ultimately guided optimization to the lead diaminothiazole compound MC180295 (14e). We then solved the crystal structure of the 14e/CDK9 complex, and found the resulting structure to be in excellent agreement with our corresponding comparative model. Finally, inspired by these models, we demonstrate how structural changes to 14e can be used to rationally tune this compounds selectivity profile. With the emergence of CDK9 as a promising target for therapeutic intervention in cancer, we anticipate that comparative modeling can provide a valuable tool to guide optimization of potency and selectivity of new inhibitors targeting this kinase.

cancer biology↗

bioRxiv2020

Rationalizing PROTAC-mediated ternary complex formation using Rosetta

PROTACs are molecules that combine a target-binding warhead with an E3 ligase-recruiting moiety; by drawing the target protein into a ternary complex with the E3 ligase, PROTACs induce target protein degradation. While PROTACs hold exciting potential as chemical probes and as therapeutic agents, development of a PROTAC typically requires synthesis of numerous analogs to thoroughly explore variations on the chemical linker; without extensive trial and error, it is unclear how to link the two protein-recruiting moieties to promote formation of a productive ternary complex. Here, we describe a structure-based computational method for evaluating suitability of a given linker for ternary complex formation. Our method uses Rosetta to dock the protein components, then builds the PROTAC from its component fragments into each binding mode; complete models of the ternary complex are then refined. We apply this approach to retrospectively evaluate multiple PROTACs from the literature, spanning diverse target proteins. We find that modeling ternary complex formation is sufficient to explain both activity and selectivity reported for these PROTACs, implying that other cellular factors are not key determinants of activity in these cases. We further find that interpreting PROTAC activity is best approached using an ensemble of structures of the ternary complex rather than a single static conformation, and that members of a structurally-conserved protein family can be recruited by the same PROTAC through vastly different binding modes. To encourage adoption of these methods and promote further analyses, we disseminate both the computational methods and the models of ternary complexes. Significance StatementRecent years have brought a flood of interest in developing compounds that selectively degrade protein targets in cells, as exemplified by PROTACs. Fully realizing the promise of PROTACs to transform chemical biology by delivering degraders of diverse and undruggable protein targets has been impeded, however, by the fact that designing a suitable chemical linker between the functional moieties requires extensive trial and error. Here, we describe a structure-based computational method to predict PROTAC activity. We envision that this approach will allow design and optimization of PROTACs for efficient target degradation, selection of E3 ligases best suited for pairing with a given target protein, and understanding the basis by which PROTACs can exhibit different target selectivity than their component warheads.

biochemistry↗