It has been previously shown that human cardiac disorders can be modeled with induced pluripotent stem cell differentiated cardiomyocytes (iPSC-CM), which enables to study disease characteristics and pathophysiology in more detail. We have shown that some genetic cardiac diseases can be separated from each other and from healthy controls by applying machine learning methods to calcium transient signals measured from these cells. In this study, separation of four genetic cardiac diseases and controls were studied by applying classification methods such as nearest neighbor searching algorithm, decision trees, least squares support vector machines and random forests to peak data computed from calcium transient signals measured from beating induced pluripotent stem cell-derived (iPSC) cardiomyocytes. The best classification accuracy obtained was 77.8% being very promising. The result strengthens our previous finding that the machine learning method can be exploited to identification of several genetic cardiac diseases, but also to separate mutations in different genes resulting in the same clinical phenotype.