Abstract
Atrioventricular septal defect (AVSD) is a clinically significant subtype of congenital heart disease (CHD) that severely influences the health of babies during birth and is associated with Down syndrome (DS). Thus, exploring the differences in functional genes in DS samples with and without AVSD is a critical way to investigate the complex association between AVSD and DS. In this study, we present a computational method to distinguish DS patients with AVSD from those without AVSD using the newly proposed self-normalizing neural network (SNN). First, each patient was encoded by using the copy number of probes on chromosome 21. The encoded features were ranked by the reliable Monte Carlo feature selection (MCFS) method to obtain a ranked feature list. Based on this feature list, we used a two-stage incremental feature selection to construct two series of feature subsets and applied SNNs to build classifiers to identify optimal features. Results show that 2737 optimal features were obtained, and the corresponding optimal SNN classifier constructed on optimal features yielded a Matthew’s correlation coefficient (MCC) value of 0.748. For comparison, random forest was also used to build classifiers and uncover optimal features. This method received an optimal MCC value of 0.582 when top 132 features were utilized. Finally, we analyzed some key features derived from the optimal features in SNNs found in literature support to further reveal their essential roles.
Highlights
Congenital heart disease (CHD) is a defect of the heart structure at birth [1] and is one of the most common birth defects in America with 8 out 1000 newborns affected with different severities [2].Genes 2018, 9, 208; doi:10.3390/genes9040208 www.mdpi.com/journal/genesIn 2010, more than 35,000 babies in the United States are born with congenital heart disease (CHD) [2,3]
self-normalizing neural network (SNN) open the door for deep network applications on general data and are not limited to sequential and image data. It has been evaluated on 121 UCI tasks, and the results reveal that SNNs are superior to forward neural network (FNN) in all tasks and have outperformed random forests (RFs) and support vector machine (SVM) when the data size is greater than 1000
Similar to SNN algorithm, we applied RF algorithm to build classifiers on feature subsets derived from incremental feature selection (IFS) method, and each classifier was evaluated by a 10-fold cross validation test
Summary
Congenital heart disease (CHD) is a defect of the heart structure at birth [1] and is one of the most common birth defects in America with 8 out 1000 newborns affected with different severities [2].Genes 2018, 9, 208; doi:10.3390/genes9040208 www.mdpi.com/journal/genesIn 2010, more than 35,000 babies in the United States are born with CHD [2,3]. Congenital heart disease (CHD) is a defect of the heart structure at birth [1] and is one of the most common birth defects in America with 8 out 1000 newborns affected with different severities [2]. More than 1 million adults live with CHD in the United States [3], implying that this disease is a widely distributed and significant threat to the health of human beings. As a group of heart structural problems at birth, CHD can be further categorized into various subtypes based on detailed pathogenesis, including different regions of the cardiovascular system and similar symptoms [6]. Approximately 15% to 20% of newborns with DS suffer from the AVSD, indicating the potential relationship between these two diseases [10,11]
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.
