Covalent drug discovery is gaining increasing interest and clinical success, especially for cancer and infectious diseases. Although old covalent drugs were discovered serendipitously, modern covalent drug discovery takes a rational approach. While chemical intuition and trial-and-error are helpful, the discovery process can be sig- ni?cantly shortened by the knowledge of whether the intended engagement site is suf?ciently nucleophilic and whether alternative reactive sites are available. Another gap that the proposal seeks to ?ll is the lack of com- mercial tools for accurate and reliable prediction of protonation states of proteins and other macromolecules. Protonation states of proteins can change upon binding, and incorrect assignment of protonation states can lead to erroneous docking, scoring results, and inaccurate binding af?nity calculations. Knowledge of protonation states is particularly relevant for drug design involving pH-sensitive targets, such as the highly pursued proteases and kinases, which often change protonation states upon activation or deactivation. The objective of this project is to develop a cloud-based web application to predict reactive hotspots in proteins and more broadly the pKa values of any titratable sites in macromolecules.
In Aim 1, we plan to further improve and validate continuous constant pH molecular dynamics for predicting pKa's and reactive hotspots in proteins.
In Aim 2, we plan to develop a on-demand web application based on the continuous constant pH molecular dynamics tool and cloud-computing paradigm for assisting drug and materials design. This unique product can be used to assist the discovery of novel covalent inhibitors, identi?cation of new drug targets, and mechanistic studies of covalent inhibition. Additionally, it can be incorporated in the current work?ow of structure-based drug design to improve the outcomes of docking, scoring, binding af?nity calculations, and molecular dynamics simulations.
Covalent inhibition is becoming a widely pursued alternative therapeutic strategy for cancer, infectious, and im- munological diseases. This project seeks to develop a cloud-based on-demand web application that offers rapid and accurate predictions of covalently targetable hotspots in proteins and more broadly the protonation states of any titratable sites in macromolecules to accelerate drug discovery.
