Composite Biomarkers Derived from Micro-Electrode Array Measurements and Computer Simulations Improve the Classification of Drug-Induced Channel Block

Abstract

The Micro-Electrode Array (MEA) device enables high-throughput electrophysiology measurements that are less labor-intensive than patch-clamp based techniques. Combined with human-induced pluripotent stem cells cardiomyocytes (hiPSC-CM), it represents a new and promising paradigm for automated and accurate in vitro drug safety evaluation. In this article, the following question is addressed: which features of the MEA signals should be measured to better classify the effects of drugs? A framework for the classification of drugs using MEA measurements is proposed. The classification is based on the ion channels blockades induced by the drugs. It relies on an in silico electrophysiology model of the MEA, a feature selection algorithm and automatic classification tools. An in silico model of the MEA is developed and is used to generate synthetic measurements. An algorithm that extracts MEA measurements features designed to perform well in a classification context is described. These features are called composite biomarkers. A state-of-the-art machine learning program is used to carry out the classification of drugs using experimental MEA measurements. The experiments are carried out using five different drugs: mexiletine, flecainide, diltiazem, moxifloxacin, and dofetilide. We show that the composite biomarkers outperform the classical ones in different classification scenarios. We show that using both synthetic and experimental MEA measurements improves the robustness of the composite biomarkers and that the classification scores are increased.

Identifying drug response by combining measurements of the membrane potential, the cytosolic calcium concentration, and the extracellular potential in microphysiological systems

“…Time resolution can be improved but at the cost of less accurate data and therefore another experimental technique is needed to pin down the channel density of and drug effects on I Na . It is well known that the extracellular potential can be measured in microphysical systems using multielectrode arrays; see e.g., [12,13,14,15,16]. In this report we will show that the extracellular data can be used to determine the sodium current.…”

“…From the Ca-trace, we include a number of CaD-values (see (10)) in addition to the Ca transient time to peak, CaP (see (13)). From the U data, we include the value in each time point and each electrode (see (14)) and the conduction velocity computed using the time points in which U crosses 0 mV after the peak in different electrodes (see (15) and Figure 1). of a * marking the data values is used for all the terms in the cost function (see below).…”

“…Considering the cost function terms, H j , defined above, all the terms may straightforwardly be defined on the form (17) except for the term H U defined in (14). For this term, we instead definē…”

“…where H 0 U is defined by (14) with u k, * e replaced by the model solution u k e (λ 0 ), and H U (λ) is defined by (14).…”

“…[17]) for this purpose. The bidomain model has already been used for inversion of U data (but not V and Ca) by several authors; see [18,16,19,20,14]. However, it is demonstrated in [18] that the bidomain model does not provide an extracellular repolarization wave.…”