Dissertation/Thesis Abstract

An Efficient Method to Globally Identify Nonlinearly Interacting Inputs to Biological Systems
by Fulton-Howard, Brian, Ph.D., Icahn School of Medicine at Mount Sinai, 2017, 100; 10616808
Abstract (Summary)

Finding interactions between inputs to complex biological systems is a question of major interest and great challenges. Biological inputs can often behave differently when combined with other inputs than alone because of biological complexity. The complexity of organisms arises from multiple overlapping mechanisms for interacting biological processes, and complex amplification, feedback and feedforward effects. This leads to complex interactions between stimuli, neurotransmitters, hormones and drugs. Understanding the interactions between natural inputs like endogenous compounds can help us understand normal and diseased biological mechanisms, and understanding interactions between drugs can lead to greater safety and efficacy with the use of multiple drugs. Various theoretical methods exist to predict interactions but they are biased by current knowledge. Methods to experimentally find interactions are handicapped by combinatorial complexity or do not take important dose dependent effects into account. Here we present Combinatorial Reduction of Interacting Tuple Tests (CRITT), an unbiased experimental method to identify biological nonlinearity. CRITT uses a heuristic to pool inputs at multiple doses, then uses algorithms to automatically deconvolve the pools and identify nonlinear pairs. We also demonstrate the efficacy of CRITT on a model of the Guinea Pig ventricular myocyte. CRITT can be expected to reduce the number of required experiments by at least two thirds.

Indexing (document details)
Advisor: Sobie, Eric A., Salton, Stephen
Commitee: Brezina, Vladimir, Dumitriu, Dani, Proekt, Alexander, Shapiro, Matthew
School: Icahn School of Medicine at Mount Sinai
Department: Neurosciences
School Location: United States -- New York
Source: DAI-B 79/01(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Biostatistics, Pharmacology, Biophysics
Keywords: Combinatorics, Nonlinear, Pairwise, Pooling, Synergy, Techniques
Publication Number: 10616808
ISBN: 9780355147889
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest