Dataset Information

Characterizing arrhythmia using machine learning analysis of Ca²⁺ cycling in human cardiomyocytes.

ABSTRACT: Accurate modeling of the heart electrophysiology to predict arrhythmia susceptibility remains a challenge. Current electrophysiological analyses are hypothesis-driven models drawing conclusions from changes in a small subset of electrophysiological parameters because of the difficulty of handling and understanding large datasets. Thus, we develop a framework to train machine learning classifiers to distinguish between healthy and arrhythmic cardiomyocytes using their calcium cycling properties. By training machine learning classifiers on a generated dataset containing a total of 3,003 healthy derived cardiomyocytes and their various arrhythmic states, the multi-class models achieved >90% accuracy in predicting arrhythmia presence and type. We also demonstrate that a binary classifier trained to distinguish cardiotoxic arrhythmia from healthy electrophysiology could determine the key biological changes associated with that specific arrhythmia. Therefore, machine learning algorithms can be used to characterize underlying arrhythmic patterns in samples to improve in vitro preclinical models and complement current in vivo systems.

SUBMITTER: Pang JKS

PROVIDER: S-EPMC9391413 | biostudies-literature | 2022 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Characterizing arrhythmia using machine learning analysis of Ca<sup>2+</sup> cycling in human cardiomyocytes.

Pang Jeremy K S JKS Chia Sabrina S Zhang Jinqiu J Szyniarowski Piotr P Stewart Colin C Yang Henry H Chan Woon-Khiong WK Ng Shi Yan SY Soh Boon-Seng BS

Stem cell reports 20220714 8

Accurate modeling of the heart electrophysiology to predict arrhythmia susceptibility remains a challenge. Current electrophysiological analyses are hypothesis-driven models drawing conclusions from changes in a small subset of electrophysiological parameters because of the difficulty of handling and understanding large datasets. Thus, we develop a framework to train machine learning classifiers to distinguish between healthy and arrhythmic cardiomyocytes using their calcium cycling properties. ...[more]

PMID: 35839773

Similar Datasets

Project description:AimsPhospholamban (PLB) is the key regulator of the cardiac Ca2+ pump (SERCA2a)-mediated sarcoplasmic reticulum Ca2+ stores. We recently reported that PLB is highly concentrated in the nuclear envelope (NE) from where it can modulate perinuclear Ca2+ handling of the cardiomyocytes (CMs). Since inositol 1,4,5-trisphosphate (IP3) receptor (IP3R) mediates nuclear Ca2+ release, we examined whether the nuclear pool of PLB regulates IP3-induced nuclear Ca2+ handling.Methods and resultsFluo-4 based confocal Ca2+ imaging was performed to measure Ca2+ dynamics across both nucleus and cytosol in saponin-permeabilized CMs isolated from wild-type (WT) or PLB-knockout (PLB-KO) mice. At diastolic intracellular Ca2+ ([Ca2+]i = 100 nM), the Fab fragment of the monoclonal PLB antibody (anti-PLB Fab) facilitated the formation and increased the length of spontaneous Ca2+ waves (SCWs) originating from the nuclear region in CMs from WT but not from PLB-KO mice. We next examined nuclear Ca2+ activities at basal condition and after sequential addition of IP3, anti-PLB Fab, and the IP3R inhibitor 2-aminoethoxydiphenyl borate (2-APB) at a series of [Ca2+]i. In WT mice, at 10 nM [Ca2+]i where ryanodine receptor (RyR2) based spontaneous Ca2+ sparks rarely occurred, IP3 increased fluorescence amplitude (F/F0) of overall nuclear region to 1.19 ± 0.02. Subsequent addition of anti-PLB Fab significantly decreased F/F0 to 1.09 ± 0.02. At 50 nM [Ca2+]i, anti-PLB Fab not only decreased the overall nuclear F/F0 previously elevated by IP3, but also increased the amplitude and duration of spark-like nuclear Ca2+ release events. These nuclear Ca2+ releases were blocked by 2-APB. At 100 nM [Ca2+]i, IP3 induced short SCWs originating from nucleus. Anti-PLB Fab transformed those short waves into long SCWs with propagation from the nucleus into the cytosol. In contrast, neither nuclear nor cytosolic Ca2+ dynamics was affected by anti-PLB Fab in CMs from PLB-KO mice in all these conditions. Furthermore, in WT CMs pretreated with RyR2 blocker tetracaine, IP3 and anti-PLB Fab still increased the magnitude of nuclear Ca2+ release but failed to regenerate SCWs. Finally, anti-PLB Fab increased low Ca2+ affinity mag-fluo 4 fluorescence intensity in the lumen of NE of nuclei isolated from WT but not in PLB-KO mice.ConclusionPLB regulates nuclear Ca2+ handling. By increasing Ca2+ uptake into lumen of the NE and perhaps other perinuclear membranes, the acute reversal of PLB inhibition decreases global Ca2+ concentration at rest in the nucleoplasm, and increases Ca2+ release into the nucleus, through mechanisms involving IP3R and RyR2 in the vicinity.

Dataset Information

Characterizing arrhythmia using machine learning analysis of Ca²⁺ cycling in human cardiomyocytes.

Publications

Characterizing arrhythmia using machine learning analysis of Ca<sup>2+</sup> cycling in human cardiomyocytes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Dataset Information

Characterizing arrhythmia using machine learning analysis of Ca2+ cycling in human cardiomyocytes.

Publications

Characterizing arrhythmia using machine learning analysis of Ca<sup>2+</sup> cycling in human cardiomyocytes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Characterizing arrhythmia using machine learning analysis of Ca²⁺ cycling in human cardiomyocytes.