Separation of HCM and LQT Cardiac Diseases with Machine Learning of Ca²⁺ Transient Profiles

Abstract

Background Modeling human cardiac diseases with induced pluripotent stem cells not only enables to study disease pathophysiology and develop therapies but also, as we have previously showed, it can offer a tool for disease diagnostics. We previously observed that a few genetic cardiac diseases can be separated from each other and healthy controls by applying machine learning to Ca²⁺ transient signals measured from iPSC-derived cardiomyocytes (CMs).

Objectives For the current research, 419 hypertrophic cardiomyopathy (HCM) transient signals and 228 long QT syndrome (LQTS) transient signals were measured. HCM signals included data recorded from iPSC-CMs carrying either α-tropomyosin, i.e., TPM1 (HCMT) or MYBPC3 or myosin-binding protein C (HCMM) mutation and LQTS signals included data recorded from iPSC-CMs carrying potassium voltage-gated channel subfamily Q member 1 (KCNQ1) mutation (long QT syndrome 1 [LQT1]) or KCNH2 mutation (long QT syndrome 2 [LQT2]). The main objective was to study whether and how effectively HCMM and HCMT can be separated from each other as well as LQT1 from LQT2.

Methods After preprocessing those Ca²⁺ signals where we computed peak waveforms we then classified the two mutations of both disease pairs by using several different machine learning methods.

Results We obtained excellent classification accuracies of 89% for HCM and even 100% for LQT at their best.

Conclusion The results indicate that the methods applied would be efficient for the identification of these genetic cardiac diseases.

Keywords

calcium transient signal - genetic cardiac diseases - machine learning - mutations

• References

• 1 Moretti A, Bellin M, Welling A. , et al. Patient-specific induced pluripotent stem-cell models for long-QT syndrome. N Engl J Med 2010; 363 (15) 1397-1409
• 2 Matsa E, Rajamohan D, Dick E. , et al. Drug evaluation in cardiomyocytes derived from human induced pluripotent stem cells carrying a long QT syndrome type 2 mutation. Eur Heart J 2011; 32 (08) 952-962
• 3 Lahti AL, Kujala VJ, Chapman H. , et al. Model for long QT syndrome type 2 using human iPS cells demonstrates arrhythmogenic characteristics in cell culture. Dis Model Mech 2012; 5 (02) 220-230
• 4 Kuusela J, Kim J, Räsänen E, Aalto-Setälä K. The effects of pharmacological compounds on beat rate variations in human long QT-syndrome cardiomyocytes. Stem Cell Rev Rep 2016; 12 (06) 698-707
• 5 Ojala M, Prajapati C, Pölönen RP. , et al. Mutation-specific phenotypes in hiPSC-derived cardiomyocytes carrying either myosin-binding protein C or α-tropomyosin mutation for hypertrophic cardiomyopathy. Stem Cells Int 2016; 2016: 1684792
• 6 Han L, Li Y, Tchao J. , et al. Study familial hypertrophic cardiomyopathy using patient-specific induced pluripotent stem cells. Cardiovasc Res 2014; 104 (02) 258-269
• 7 Lan F, Lee AS, Liang P. , et al. Abnormal calcium handling properties underlie familial hypertrophic cardiomyopathy pathology in patient-specific induced pluripotent stem cells. Cell Stem Cell 2013; 12 (01) 101-113
• 8 Penttinen K, Swan H, Vanninen S. , et al. Antiarrhythmic effects of dantrolene in patients with catecholaminergic polymorphic ventricular tachycardia and replication of the responses using iPSC models. PLoS One 2015; 10 (07) e0134746
• 9 Juhola M, Joutsijoki H, Penttinen K, Aalto-Setälä K. Detection of genetic cardiac diseases by Ca²⁺ transient profiles using machine learning methods. Sci Rep 2018; 8 (01) 9355
• 10 Maron BJ, Ommen SR, Semsarian C, Spirito P, Olivotto I, Maron MS. Hypertrophic cardiomyopathy: present and future, with translation into contemporary cardiovascular medicine. J Am Coll Cardiol 2014; 64 (01) 83-99
• 11 Wang Q, Curran ME, Splawski I. , et al. Positional cloning of a novel potassium channel gene: KVLQT1 mutations cause cardiac arrhythmias. Nat Genet 1996; 12 (01) 17-23
• 12 Sanguinetti MC, Jiang C, Curran ME, Keating MT. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the IKr potassium channel. Cell 1995; 81 (02) 299-307
• 13 Lee EK, Tran DD, Keung W. , et al. Machine learning of human pluripotent stem cell-derived engineered cardiac tissue contractility for automated drug classification. Stem Cell Rep 2017; 9 (05) 1560-1572
• 14 Juhola M, Joutsijoki H, Varpa K. , et al. On computation of calcium cycling anomalies in cardiomyocytes data. Paper presented at: Annual International Conference IEEE Engineering in Medicine and Biology Society, 2014; Chicago, IL; 1444–1447
• 15 Juhola M, Penttinen K, Joutsijoki H. , et al. Signal analysis and classification methods for the calcium transient data of stem cell-derived cardiomyocytes. Comput Biol Med 2015; 61: 1-7
• 16 Juhola M, Joutsijoki H, Penttinen K, Aalto-Setälä K. Machine learning to differentiate diseased cardiomyocytes from healthy control cells. Inf Med Unlocked 2019; 14: 15-22
• 17 Mummery C, Ward-van Oostwaard D, Doevendans P. , et al. Differentiation of human embryonic stem cells to cardiomyocytes: role of coculture with visceral endoderm-like cells. Circulation 2003; 107 (21) 2733-2740
• 18 Kujala K, Paavola J, Lahti A. , et al. Cell model of catecholaminergic polymorphic ventricular tachycardia reveals early and delayed afterdepolarizations. PLoS One 2012; 7 (09) e44660
• 19 van der Maaten LJP. Accelerating t-SNE using tree-based algorithms. J Mach Learn Res 2015; 15: 3221-3245
• 20 van der Maaten LJP, Hinton GE. Visualizing high-dimensional data using t-SNE. J Mach Learn Res 2008; 5: 2579-2605
• 21 Thanh Noi P, Kappas M. Comparison of random forest, k-nearest neighbor, and support vector machine classifiers for land cover classification using Sentinel-2 imagery. Sensors (Basel) 2017; 18 (01) 18
• 22 Dudani SA. The distance weighted k-nearest neighbor rule. IEEE Trans Syst Man Cybern 1976; 6 (04) 325-327
• 23 Tharwat A, Gaber T, Ibrahim A, Hassanien AE. Linear discriminant analysis: a detailed tutorial. AI Commun 2017; 30 (02) 169-190
• 24 Bohling G. Classical normal-based discriminant analysis. Technical Report EECS 2006; 833: 1-24 . Available at: http://people.ku.edu~gbohling/EECS833/Discrim.pdf
• 25 Tharwat A. Linear vs. quadratic discriminant analysis classifier: a tutorial. Int J Applied Pattern Recognit 2016; 3 (02) 145-179
• 26 Gokgoz E, Subasi A. Comparison of decision tree algorithms for EMG signal classification using DWT. Biomed Signal Process Control 2015; 18: 138-144
• 27 Kwak C, Clayton-Matthews A. Multinomial logistic regression. Nurs Res 2002; 51 (06) 404-410
• 28 Escalante HJ, Morales EF, Sucar E. A naïve Bayes baseline for early gesture recognition. Pattern Recognit Lett 2016; 73: 91-99
• 29 Liu JNK, He YL, Wang XZ, Hu YX. A comparative study among different kernel functions in flexible naïve Bayesian classification. Paper presented at: Proceedings of 2011 International Conference on Machine Learning Cybernetics; 2011: 638-643
• 30 Breiman L. Random forests. Mach Learn 2001; 45 (01) 5-32
• 31 Biau G, Scornet E. A random forest guided tour. Test 2016; 25 (02) 197-227
• 32 Suykens JAK, Vandewalle J. Least squares support vector machine classifiers. Neural Process Lett 1999; 9 (03) 293-300
• 33 Yang A, Li W, Yang X. Short-term electricity load forecasting based on feature selection and least squares support vector machines. Knowl Base Syst 2019; 163: 159-173