Development of Feature-Based Receptor-Ligand Docking Using POVME
CSEF · 2015 Biochemistry/ Molecular Biology
Overview
Objectives/Goals The objective of this work was to develop an efficient receptor-ligand docking method integrated into a binding pocket analysis tool (POVME) that can potentially be used to increase the rate of drug discovery. Two major goals include the generation of uniform rotations in 3-D space and the evaluation of the accuracy of the docking and scoring functions in determining the correct position of a ligand within a receptor. Methods/Materials A computer with access to the Binding Database (a public, web-accessible database of measured binding affinities, focusing chiefly on the interactions of proteins considered to be drug-targets with small, drug-like molecules) was used to write and test coloring, docking, and scoring functions. Functions were written in Python and C++. Visualizations were produced using VMD. Results When testing methods for generating uniform rotations, point-repulsion-generated points on a sphere had the lowest standard deviation between point distances. Testing of the docking and scoring functions showed that there existed a strong negative linear association between feature scores and distance from correct ligand binding position. Conclusions/Discussion The low standard deviation of distances between points generated by the point repulsion method helped to fulfill design goal #1, producing more uniform and regular rotations in 3-D space. In addition, due to the strong negative linear association between feature interaction score and distance from the correct ligand orientation position, there is moderate evidence supporting that design goal #2 has been met. Overall, results show that the docking function, in a naïve case, produces scores that increase as the distance from the true ligand conformation decreases. This data suggests that the docking and scoring functions have the potential to reliably select the true ligand conformations of any given receptor and ligand maps. These advances allow accurate docking and scoring to be implemented within POVME.
Summary statement
This work introduced binding-pocket analysis-based receptor-ligand docking and optimized scoring functions in order to expedite future drug discovery.
Help received
Participant in BioChemCoRe program at University of San Diego, California; continued research using lab equipment from the Amaro Lab under the supervision of Mr. Jeffrey R. Wagner.
Competition history
- CSEF 2015
Resources
Related projects
ISEF · 2015
Rethinking Drug Discovery: New Algorithms for Virtual Drug Screening
ISEF · 2016
A New Method for Simulation-Aided Drug Discovery Using Machine Learning Algorithms
ISEF · 2025
A Novel Computational Generative Model That Identifies Ligands: A Potential Way to Reduce Cost & Time in Identifying Promising Drug Candidates
ISEF · 2017
Development of a Software Tool for Protein-Protein Complex Prediction
CSEF · 2012
Modeling and Molecular Dynamics Simulations of Membrane-Bound Aromatase Reveal Novel Druggable Sites
ISEF · 2018
Computational Predictions in the Design of Affinity-Based Drug Delivery
CSEF · 2016
Quantifying the Complexity of Conformational Transitions in the Partial and Biased Activations of GPCRs
CSEF · 2013
Novel Software Tool for Structural Analysis of MHC Interactions with Immune Epitopes
Closest projects by meaning, across every fair and year in the corpus.
Browse more like this
Source: California Science & Engineering Fair public projects