Abstract
SummaryHuman induced pluripotent stem cell-derived cardiomyocytes have been established to detect dynamic calcium transients by fast kinetic fluorescence assays that provide insights into specific aspects of clinical cardiac activity. However, the precise derivation and use of waveform parameters to predict cardiac activity merit deeper investigation. In this study, we derived, evaluated, and applied 38 waveform parameters in a novel Python framework, including (among others) peak frequency, peak amplitude, peak widths, and a novel parameter, shoulder-tail ratio. We then trained a random forest model to predict cardiac activity based on the 25 parameters selected by correlation analysis. The area under the curve (AUC) obtained for leave-one-compound-out cross-validation was 0.86, thereby replicating the predictions of conventional methods and outperforming fingerprint-based methods by a large margin. This work demonstrates that machine learning is able to automate the assessment of cardiovascular liability from waveform data, reducing any risk of user-to-user variability and bias.
Highlights
We developed a Python toolkit, CardioWave, to derive 38 parameters from calcium transient waveforms of 63 unique compounds, using data provided by AstraZeneca (AZ) and GlaxoSmithKline (GSK)
In addition to conventional parameters such as peak frequency and amplitude-related parameters, we defined and calculated for all compounds several novel parameters, such as shoulder-tail ratio and the presence of multiple peaks. We demonstrated that these parameters can be used in combination with machine learning models to flag potential cardiac activity later in the clinic with a performance comparable to alternative approaches, but while automating the process and removing individual bias from it
Many studies have shown the usefulness of calcium transients in hiPSC-CMs as a high-throughput technique to detect the potential risk of cardiovascular toxicity by assessment of drug-induced long QT and torsades de pointes (TdP) risk or cardiomyocyte contraction (Kopljar et al, 2018; Lu et al, 2019; Pointon et al, 2015)
Summary
Cardiovascular toxicity often results in delays in drug discovery and development and additional clinical monitoring (Cook et al, 2014). For the past 25 years, the focus of cardiotoxicity assessments has centered around human ether-a-go-go-related gene (hERG)-mediated QT prolongation as a predictor of pro-arrhythmia risk (Wisniowska et al, 2014). Cardiovascular toxicity can be the result of changes in hemodynamics (i.e., heart rate, blood pressure, and cardiac contractility) and/or changes in cardiovascular pathology. These elements are undetected via the assessment of pro-arrhythmia (Mellor et al, 2011). Several in vitro approaches have been developed recently to assess the potential of new chemical entities (NCEs) to induce structural and functional changes in cardiomyocytes. Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) have in particular been widely used in recent years, as they present an opportunity to consistently generate normal patient- and disease-specific cell lines (Gintant et al, 2016; Hoekstra et al, 2012)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
