What does the system care about? Empirical approaches to identifying biological regulation

by Prof. Naama Brenner

Dept. of Chemical Engineering, Technion – Israel Institute of Technology

at Biological and soft-matter physics

Thu, 01 Feb 2024, 12:10
Sacta-Rashi Building for Physics (54), room 207, and ZOOM hybrid ID: 895 7808 4386, Passcode: 345123

Abstract

The complexity of biological regulation makes it difficult to understand, model or even identify what quantity is under regulation. Here two approaches are presented to empirically identify regulated variables in experimental data.

First, a heuristic approach is employed to the well-studied process of bacterial growth and division. Robustness against perturbations is used as a measure for regulation, and a hierarchy of variables are identified that span a range of regulation. These results are manifested geometrically as a control manifold in the space of variables: set points span a wide range of values within the manifold, whereas out-of-manifold deviations are constrained.

Then, a more general computational framework is presented, utilizing the strength of machine learning methods. Defining a coefficient of regulation enables one to set up an optimization problem mathematically. We designed a two-player algorithm that solves this nontrivial optimization problem and identifies the most conserved combination of variables. The algorithm developed shows remarkable success in benchmark validations and can be applied to various data-sets.

References: Stawsky et al., iScience 2022. Teichner et al., PNAS 2023.

For hybrid: Join Zoom Meeting
https://us02web.zoom.us/j/89578084386?pwd=MkwyRmhwVnMwUzdYMHpaMHg4UXlHUT09

Meeting ID: 895 7808 4386
Passcode: 345123

Created on 28-12-2023 by Granek, Rony (rgranek)
Updaded on 01-02-2024 by Granek, Rony (rgranek)