Type

Text

Type

Dissertation

Advisor

Rizzo, Robert C | Green, David | Garcia-Diaz, Miguel | Ojima, Iwao

Date

2017-12-01

Keywords

denovo design | Cheminformatics | Bioinformatics | docking | hivgp41 | molecular modeling

Department

Department of Biochemistry and Structural Biology

Language

en_US

Source

This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.

Identifier

http://hdl.handle.net/11401/78224

Publisher

The Graduate School, Stony Brook University: Stony Brook, NY.

Format

application/pdf

Abstract

In this dissertation, we discuss the implementation and evaluation of several improvements to the DOCK6 codebase relevant to the development of small molecule HIVgp41 inhibitors. In Chapter1, we give a brief introduction for the dissertation. This includes a description of the mechanism of HIV fusion, and a summary of existing FDA approved HIV inhibitors along with a history of not yet approved gp41 fusion inhibitors. Further, this chapter introduces the computational techniques of docking and molecular dynamics which are utilized throughout this work. In Chapter 2, we describe the development and validation of a de novo design protocol for DOCK6. The protocol builds novel small molecules from a library of fragments specifically tailored to protein target and a user defined set of scoring criteria and molecular properties. The chapter goes into great depth on the library generation and fragment assembly routines. Additionally, validation is provided on 657 ‘focused’ rebuilding tests, and 57 ‘generic’ large scale tests. Finally, the method is applied to HVgp41 to generate prospective leads. It is anticipated that this method will supplement existing lead discovery efforts. Chapter 3, explores the small molecule torsional sampling routines employed by DOCK6 identifying inadequacies in past sampling methods and proposing two improvements which increase the accuracy of predicted binding poses. In Chapter 4, we seek to improve the activity of several previously discovered HIVgp41 inhibitors. Manual refinement is performed to strengthen interactions between the inhibitor and the target which mimic interactions between the target and the native substrate. Further, the de novo protocol, described in Chapter 2 is used to automate sampling of ‘R’ groups. Chapter 5, explores the impact of receptor structure selection on enrichment outcomes and evaluates the potential of two protocols MAR and MIR to model receptor flexibility in a computationally efficient manner for docking. In Chapter 6 we summarize the work presented in the dissertation, describing the scientific impact, challenges, and future studies to further aid de novo design and HIVgp41 inhibition. | 167 pages

Download

Share

COinS