Preventive SNP–SNP interactions in the mitochondrial displacement loop (D-loop) from chronic dialysis patients Jin-Bor Chen a , Li-Yeh Chuang b , Yu-Da Lin c , Chia-Wei Liou d , Tsu-Kung Lin d , Wen-Chin Lee a , Ben-Chung Cheng a , Hsueh-Wei Chang e, ⁎, Cheng-Hong Yang c, ⁎⁎ a Division of Nephrology, Department of Internal Medicine, Mitochondrial Research Unit, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Kaohsiung, Taiwan b Department of Chemical Engineering & Institute of Biotechnology and Chemical Engineering, I-Shou University, Kaohsiung, Taiwan c Department of Electronic Engineering, National Kaohsiung University of Applied Sciences, Kaohsiung, Taiwan d Department of Neurology and Mitochondrial Research Unit, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Kaohsiung, Taiwan e Department of Biomedical Science and Environmental Biology, Cancer Center, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan abstract article info Article history: Received 5 November 2012 Received in revised form 24 January 2013 Accepted 31 January 2013 Available online xxxx Keywords: Mitochondrial DNA disease SNP interaction SNP barcode Chronic dialysis D-loop Protective Algorithm Chronic dialysis association study involving individual single nucleotide polymorphisms (SNPs) in the mitochon- drial displacement loop (D-loop) has previously been reported. However, possible SNP–SNP interactions for SNPs in the D-loop which could be associated with a reduced risk for chronic dialysis were not investigated. The purpose of this study was to propose an effective algorithm to identify protective SNP–SNP interactions in the D-loop from chronic dialysis patients. We introduce ISGA that uses an initialization strategy for genetic algo- rithms (GA) to improve the computational analysis for protective SNP–SNP interactions. ISGA generates geno- type patterns with combined SNPs (SNP barcodes) for chronic dialysis. Using our previously reported 77 SNPs in the D-loop, the algorithm-generated protective SNP barcodes for chronic dialysis were evaluated. ISGA pro- vides the SNP barcodes with the maximum frequency differences of occurrence between the cases and controls. The identified SNP barcodes with the lowest odds ratio (OR) values were regarded as the best preventive SNP barcodes against chronic dialysis. The best ISGA-generated SNP barcodes (two to nine SNPs) are more closely as- sociated with the prevention of chronic dialysis when more SNPs are chosen (OR = 0.64 to 0.32; 95% confidence interval=0.882 to 0.198). The cumulative effects of SNP–SNP interactions were more dominant in ISGA rather than in GA without the initialization strategy. We provide a fast identification of chronic dialysis-associated pro- tective SNP barcodes and demonstrate that the SNP–SNP interactions may have a cumulative effect on prediction for chronic dialysis. © 2013 Elsevier B.V. and Mitochondria Research Society. All rights reserved. 1. Introduction Single nucleotide polymorphisms (SNPs) are widely used to evalu- ate the susceptibilities to diseases. Different SNPs may have a different impact on the occurrence of a disease. Single SNP analysis strategies are commonly used to identify SNPs with significant associations, and are not suitable for identifying SNPs associated with complex polygenic diseases, which may partly explain the “missing heritability” (Cordell, 2009; Eichler et al., 2010; Manolio et al., 2009). Accordingly, the jointed effect of the combination of individual SNPs in gene–gene interactions, or epistasis, allows evaluation of the overall risk of diseases (Rose and Bell, 2012). The importance of SNP interactions has been well documented in the genomics field (Chang et al., 2012; Lane et al., 2012; Steen, 2012). Increas- ing evidence suggests that SNP–SNP interactions may confer a cumulative association of multiple SNPs with many diseases (Heslop et al., 2012; Jung et al., 2009; Kim et al., 2012; Lin et al., 2008, 2009; Vogelsang et al., 2012; Yen et al., 2008; Zheng et al., 2008). In our previous chronic dialysis association study (Chen et al., 2012), we investigated the disease predis- position of individual SNPs without considering the SNP–SNP interac- tions. 77 SNPs in the mitochondrial displacement loop (D-loop) were reported and only nine of these SNPs constituted a significant risk. The relationship of possible SNP–SNP interactions in this previous association study remains unclear. Recently, some association hinted at a reduction of the disease risk as- sociated with certain SNPs (Hollegaard et al., 2013; Qiao et al., 2008; Ragnarsdottir et al., 2010; Rodrigues et al., 2012; Velavan et al., 2012). Other previously released chronic dialysis association studies (Friedman Mitochondrion xxx (2013) xxx–xxx ⁎ Corresponding author. Tel.: +886 7 3121101x2691; fax: +886 7 3125339. ⁎⁎ Corresponding author. E-mail addresses: chenjb1019@gmail.com (J.-B. Chen), chuang@isu.edu.tw (L.-Y. Chuang), e0955767257@yahoo.com.tw (Y.-D. Lin), cwliou@ms22.hinet.net (C.-W. Liou), tklin@adm.cgmh.org.tw (T.-K. Lin), leewenchin@gmail.com (W.-C. Lee), benzmcl@yahoo.com.tw (B.-C. Cheng), changhw@kmu.edu.tw (H.-W. Chang), chyang@cc.kuas.edu.tw (C.-H. Yang). MITOCH-00796; No of Pages 7 1567-7249/$ – see front matter © 2013 Elsevier B.V. and Mitochondria Research Society. All rights reserved. http://dx.doi.org/10.1016/j.mito.2013.01.013 Contents lists available at SciVerse ScienceDirect Mitochondrion journal homepage: www.elsevier.com/locate/mito Please cite this article as: Chen, J.-B., et al., Preventive SNP–SNP interactions in the mitochondrial displacement loop (D-loop) from chronic dialysis patients, Mitochondrion (2013), http://dx.doi.org/10.1016/j.mito.2013.01.013