Identifying drug effects in a cardiac model of electrophysiology using kernel-based parameter estimation methods

BioRxiv

bioRxiv Subject Collection: Systems Biology
This feed contains articles for bioRxiv Subject Collection "Systems Biology"

Identifying drug effects in a cardiac model of electrophysiology using kernel-based parameter estimation methods

Computational models of cardiac electrophysiology hold great potential in the drug discovery process to bridge the gap between in vitro and in vivo preclinical trials. Current methods for solving inverse problems in cardiac electrophysiology are limited by their accuracy, scalability, practicality, or a combination of these. We investigate the feasibility of using kernel methods to solve the inverse problem of estimating the parameters of ionic membrane currents from observations of corresponding action potential (AP) traces. In particular, we consider AP traces generated by a cardiac cell action potential model, which mimics those obtained experimentally in measurable in vitro cardiac systems. This proof-of-concept study aims to improve existing pipelines for identifying drug effects in "heart-on-a-chip" systems by introducing a new approach to solving the inverse problem. Using synthetic training data from the 1977 Beeler-Reuter AP model of mammalian ventricular cardiomyocytes, we demonstrate our recently proposed boosted KRR solver StreaMRAK, which is particularly robust and well-adapted for high-complexity functions. We show that this method is less memory demanding, estimates the model parameters with higher accuracy, and is less sensitive to parameter identifiability problems than existing methods, such as standard KRR solvers and loss-minimization schemes based on nearest neighbor heuristics.
Oslandsbotn, A., Forsch, N., Cloninger, A.
March 17, 2023
http://biorxiv.org/cgi/content/short/2023.03.15.532862v1?rss=1