Genetic algorithm-based personalized models of human cardiac action potential

• Published in: PloS one Vol. 15; no. 5; p. e0231695
• Main Authors: Smirnov, Dmitrii, Pikunov, Andrey, Syunyaev, Roman, Deviatiiarov, Ruslan, Gusev, Oleg, Aras, Kedar, Gams, Anna, Koppel, Aaron, Efimov, Igor R.
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 11.05.2020; Public Library of Science (PLoS)
• Subjects: Action potential; Action Potentials - genetics; Action Potentials - physiology; Algorithms; Biology and Life Sciences; Cardiomyocytes; Convergence; Customization; Electrophysiology; Engineering and Technology; Gene Expression; Gene mapping; Genes; Genetic algorithms; Heart; Heart - physiology; Heart conduction system; Heart rate; Heart Transplantation; Heart Ventricles - metabolism; Humans; Mapping; Mathematical models; Medicine and Health Sciences; Messenger RNA; Methods; Models, Biological; Mutation; Myocytes, Cardiac - physiology; Noise levels; Novels; Organisms; Parameter modification; Patch-Clamp Techniques; Physical Sciences; Physics; Precision medicine; Research and Analysis Methods; Restitution; RNA; RNA-Seq; Signal to noise ratio; Simulation; Tissue Donors; Ventricle; Waveforms; Russia; United States > US
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0231695

We present a novel modification of genetic algorithm (GA) which determines personalized parameters of cardiomyocyte electrophysiology model based on set of experimental human action potential (AP) recorded at different heart rates. In order to find the steady state solution, the optimized algorithm performs simultaneous search in the parametric and slow variables spaces. We demonstrate that several GA modifications are required for effective convergence. Firstly, we used Cauchy mutation along a random direction in the parametric space. Secondly, relatively large number of elite organisms (6-10% of the population passed on to new generation) was required for effective convergence. Test runs with synthetic AP as input data indicate that algorithm error is low for high amplitude ionic currents (1.6±1.6% for IKr, 3.2±3.5% for IK1, 3.9±3.5% for INa, 8.2±6.3% for ICaL). Experimental signal-to-noise ratio above 28 dB was required for high quality GA performance. GA was validated against optical mapping recordings of human ventricular AP and mRNA expression profile of donor hearts. In particular, GA output parameters were rescaled proportionally to mRNA levels ratio between patients. We have demonstrated that mRNA-based models predict the AP waveform dependence on heart rate with high precision. The latter also provides a novel technique of model personalization that makes it possible to map gene expression profile to cardiac function.