Identify SNP-SNP Interactions Research Articles

Distributed multi-objective optimization for SNP-SNP interaction detection

The detection of complex interactions between single nucleotide polymorphisms (SNPs) plays a vital role in genome-wide association analysis (GWAS). The multi-objective evolutionary algorithm is a promising technique for SNP-SNP interaction detection. However, as the scale of SNP data further increases, the exponentially growing search space gradually becomes the dominant factor, causing evolutionary algorithm (EA)-based approaches to fall into local optima. In addition, multi-objective genetic operations consume significant amounts of time and computational resources. To this end, this study proposes a distributed multi-objective evolutionary framework (DM-EF) to identify SNP-SNP interactions on large-scale datasets. DM-EF first partitions the entire search space into several subspaces based on a space-partitioning strategy, which is nondestructive because it guarantees that each feasible solution is assigned to a specific subspace. Thereafter, each subspace is optimized using a multi-objective EA optimizer, and all subspaces are optimized in parallel. A decomposition-based multi-objective firework optimizer (DCFWA) with several problem-guided operators was designed. Finally, the final output is selected from the Pareto-optimal solutions in the historical search of each subspace. DM-EF avoids the preference for a single objective function, handles the heavy computational burden, and enhances the diversity of the population to avoid local optima. Notably, DM-EF is load-balanced and scalable because it can flexibly partition the space according to the number of available computational nodes and problem size. Experiments on both artificial and real-world datasets demonstrate that the proposed method significantly improves the search speed and accuracy.

Investigation of Genetic Polymorphisms in BMP2, BMP4, SMAD6, and RUNX2 and Persistent Apical Periodontitis

IntroductionThis study aimed to evaluate the interplay among single-nucleotide polymorphisms (SNPs) in the encoding genes BMP2, BMP4, SMAD6, and RUNX2 in persistent apical periodontitis (PAP). MethodsIn this multicentric study, 272 patients diagnosed with pulp necrosis with apical periodontitis before root canal therapy who attended regular follow-up visits for at least 1 year were screened. Periapical radiographs and clinical aspects were evaluated, and the participants were classified as PAP (n = 110) or repaired (n = 162). Genomic DNA was used for the genotyping of the following SNPs: rs1005464 and rs235768 in bone morphogenetic protein 2 (BMP2), rs17563 in bone morphogenetic protein 4 (BMP4), rs2119261 and rs3934908 in SMAD family member 6 (SMAD6), and rs59983488 and rs1200425 in runt-related transcription factor 2 (RUNX2). The chi-square test was used to compare genotype distributions between groups. The multifactor dimensionality reduction method was applied to identify SNP-SNP interactions. The alpha for all the analysis was 5%. ResultsThe multifactor dimensionality reduction suggested the rs235768 in BMP2 and rs59983488 in RUNX2 as the best SNP-SNP interaction model (cross-validation = 10/10, testing balanced accuracy = 0.584, P = .026) followed by rs17563 in BMP4 and rs2119261 in SMAD6 (cross validation = 10/10, testing balanced accuracy = 0.580, P = .031). In the rs235768 in BMP2 and rs59983488 in RUNX2 model, the high-risk genotype was TT + TT (odds ratio = 4.36; 95% confidence interval, 0.44–42.1). In model rs17563 in BMP4 and rs2119261 in SMAD6, GG + TT (odds ratio = 2.63; 95% confidence interval, 0.71–11.9) was the high-risk genotype. ConclusionsThe interactions between rs235768 in BMP2 and rs59983488 in RUNX2 and between rs17563 in BMP4 and rs2119261 in SMAD6 are associated with PAP, suggesting that an interplay of these SNPs is involved in the higher risk of developing PAP.

DNA Repair Genes May Influence Cognitive Outcomes in Adult Patients with Hematologic Malignancies (HM) Treated with Blood or Marrow Transplantation (BMT)

Introduction: We recently demonstrated that cognitive impairment (CI) is prevalent and persistent in HM patients treated with BMT (Sharafeldin et al., J Clin Oncol, 2018). While the impact of genetic factors on cognition in the non-oncology realm is well-recognized, limited information exists in HM patients treated with BMT. We tested the hypothesis that genetic variants in blood-brain barrier (BBB), drug transporter, telomere homeostasis, neurotransmission, neural repair and DNA repair are associated with CI in HM patients treated with BMT. We previously presented preliminary results of the Discovery cohort (Sharafeldin et al., Blood, 2016); here we present additional analyses (at the gene level) on the Discovery cohort, and replication of salient findings in an independent Validation cohort.Methods: The Discovery cohort consisted of prospective, longitudinally-assessed patients undergoing autologous or allogeneic BMT for HM at a single center. Cognitive function was assessed objectively using a battery of standardized tests prior to BMT, and at 6mo, 1y, 2y, and 3y post-BMT; germline DNA was collected pre-BMT. Global Deficit Score (GDS) was used as an indicator of CI (GDS ≥0.5 indicated CI). We employed a candidate gene approach, relying on biologically-plausible hypotheses for gene selection. Standard quality control procedures yielded 985 SNPs in 68 genes with a total genotyping call rate of 99% in 277 BMT recipients. Generalized estimating equation models were fitted to examine association of individual SNPs with CI. Linkage disequilibrium-based pruning yielded 326 independent tests, with an overall p-value threshold of 1.53x10-4 using Bonferroni correction for multiple testing. We also conducted gene-level analysis considering each gene as a SNP-set using the machine learning approach of logic regression, with cross-validation, to identify SNP-SNP interactions at the gene-level (Bonferroni p ≤ 7.94x10-4). An independent Validation case-control set from the BMT Survivor Study was used to replicate significant findings from the Discovery cohort, using conditional logistic regression models. Cases (n=141) were BMT survivors who reported: i) presence of memory/learning problems; and ii) severity of memory/learning problems. Controls (n=258) were BMT survivors without cognitive problems, matched (1-2/case) on race/ethnicity, HM, and time from BMT.Results:Discovery: Among the 277 patients in the Discovery cohort, mean age at BMT was 50y, 59% were males, 69% were non-Hispanic whites and 47% received an allogeneic BMT. Male sex (OR=3.2, 95%CI, 1.5-6.7, p=0.002) and lower pre-BMT cognitive reserve (OR=4.6, 95%CI, 2.2-9.6, p <0.0001) were associated with CI. Risk of CI was increased in the younger patients (<50y) if they received TBI (age*TBI interaction OR=3.86, 95%CI, 1.0-15.1, p=0.05). Final genetic models were adjusted for relevant variables with significant independent main effects including BMT type, age at BMT, sex, race/ethnicity, and cognitive reserve. We identified significant associations between CI and BBB, telomere homeostasis and DNA repair SNPs using the single-SNP (Fig 1) and gene-level analyses. Replication: The mean age at BMT was 44y, 54% were males, 88% were non-Hispanic whites and 5% received an allogeneic BMT. Replication models were adjusted for age at BMT, sex, race, and education and income levels. Of the 69 SNPs identified for replication, DNA repair SNPs from the single-SNP analysis rs13006837 on XRCC5 (OR=5.8, 95%CI, 2.7-12.3, p=2.0 x 10-5) and rs293796 on OGG1 gene (OR=11.4, 95%CI, 4.6-28.5, p-value=1.08 x 10-6) were successfully replicated. Five SNPs on DNA repair genes from the gene-level analysis showed individual statistically-significant associations in the Validation cohort: rs238417 on ERCC2 (OR=0.48, 95%CI, 0.25-0.91, p=0.03); rs2526698 on EXO1 (OR=0.57, 95%CI: 0.33-0.98, p=0·04); rs3793906 on MGMT (OR=1.96, 95%CI: 1.2-3.3, p=0.01); rs4725016 on RPA3 (OR=0.20, 95%CI, 0.05-0.92, p=0.04); and rs1479569 on XRCC4 (OR=2.3, 95%CI, 1.3-4.1, p=0.005).Conclusion: Our study provides genetic characterization of adult HM patients at increased risk of cognitive impairment following BMT; associations with DNA repair variants were independently replicated. These findings could have implications towards the personalized management of patients undergoing BMT. DisclosuresForman:Mustang Therapeutics: Other: Licensing Agreement, Patents & Royalties, Research Funding.

Catfish Taguchi-Based Binary Differential Evolution Algorithm for Analyzing Single Nucleotide Polymorphism Interactions in Chronic Dialysis.

Single-nucleotide polymorphism (SNP)-SNP interactions are crucial for understanding the association between disease-related multifactorials for disease analysis. Existing statistical methods for determining such interactions are limited by the considerable computation required for evaluating all potential associations between disease-related multifactorials. Identifying SNP-SNP interactions is thus a major challenge in genetic association studies. This paper proposes a catfish Taguchi-based binary differential evolution (CT-BDE) algorithm for identifying SNP-SNP interactions. In the search space, the catfish effect prevents the premature convergence of the population, and the Taguchi method improves the search ability of the BDE algorithm. Hence, the proposed algorithm enables obtaining a favorable solution regarding the identification of high-order SNP-SNP interactions. Additionally, the proposed algorithm applies an effective fitness function derived from a multifactor dimensionality reduction (MDR) operation to evaluate the solutions from BDE-based algorithms. Simulated and real data sets were used to evaluate the ability of several BDE-based algorithms in identifying specific SNP-SNP interactions. We compared the fitness function derived from the MDR operation with that derived according to the difference between cases and controls, by using the different BDE-based algorithms. The results showed that the proposed CT-BDE algorithm applying the fitness function derived from the MDR operation exhibited a superior ability in identifying SNP-SNP interactions compared with the other BDE-based algorithms.

Parallelization of Logic Regression Analysis on SNP-SNP Interactions of a Crohn's Disease Dataset Model

SNP-SNP interactions have been recognized to be basically important for understanding genetic causes of complex disease traits. Logic regression is an effective methods for identifying SNP-SNP interactions associated with risk of complex disease. However, identifying SNP-SNP interactions are computationally challenging and may take hours, weeks and months to complete. Although parallel computing is a powerful method to accelerate computing time, it is arduous for users to apply this method to logic regression analyses of SNP-SNP interactions because it requires advanced programming skills to correctly partition and distribute data, control and monitor tasks across multi-core CPUs or several computers, and merge output files. In this paper, we present a novel R-library called SNPInt to automatically speed up analyses of SNP-SNP interactions of genome-wide association (GWA) studies using parallel computing without the advanced programming skills. The Crohn’s disease GWA studies dataset from the Wellcome Trust Case Control Consortium (WTCCC) that includes 4,680 individuals with 500,000 SNPs’ genotypes was analyzed using logic regression on a computer cluster to evaluate SNPInt performance. The results from SNPInt with any number of CPUs are the same as the results from non-parallel approach, and SNPInt library quite accelerated the logic regression analysis. For instance, with two hundred genes and twenty permutation rounds, the computing time was continuously decreased from 7.3 days to only 0.9 day when SNPInt applied eight CPUs. Executing analyses of SNP-SNP interactions using the SNPInt library is an effective way to boost performance, and simplify the parallelization of analyses of SNP-SNP interactions.

Chaotic particle swarm optimization for detecting SNP–SNP interactions for CXCL12-related genes in breast cancer prevention

Genome-wide association studies have revealed that many single nucleotide polymorphisms (SNPs) are associated with breast cancer, and yet the potential SNP-SNP interactions have not been well addressed to date. This study aims to develop a methodology for the selection of SNP-genotype combinations with a maximum difference between case and control groups. We propose a new chaotic particle swarm optimization (CPSO) algorithm that identifies the best SNP combinations for breast cancer association studies containing seven SNPs. Five scoring functions, that is, the percentage correct, sensitivity/specificity, positive predictive value/negative predictive value, risk ratio, and odds ratio, are provided for evaluating SNP interactions in different SNP combinations. The CPSO algorithm identified the best SNP combinations associated with breast cancer protection. Some SNP interactions in specific SNPs and their corresponding genotypes were revealed. These SNP combinations showed a significant association with breast cancer protection (P<0.05). The sensitivity and specificity of the respective best SNP combinations were all higher than 90%. In contrast to the corresponding non-SNP-SNP interaction combinations, the estimated odds ratio and risk ratio of the SNP-SNP interaction in SNP combinations for breast cancer were less than 100%. This suggests that CPSO can successfully identify the best SNP combinations for breast cancer protection. In conclusion, we focus on developing a methodology for the selection of SNP-genotype combinations with a maximum difference between case and control groups. The CPSO method can effectively identify SNP-SNP interactions in complex biological relationships underlying the progression of breast cancer.

TRM: A Powerful Two‐Stage Machine Learning Approach for Identifying SNP‐SNP Interactions

Studies have shown that interactions of single nucleotide polymorphisms (SNPs) may play an important role in understanding the causes of complex disease. We have proposed an integrated machine learning method that combines two machine-learning methods-Random Forests (RF) and Multivariate Adaptive Regression Splines (MARS)-to identify a subset of important SNPs and detect interaction patterns more effectively and efficiently. In this two-stage RF-MARS (TRM) approach, RF is first applied to detect a predictive subset of SNPs, and then MARS is used to identify the interaction patterns. We evaluated the TRM performances in four models. RF variable selection was based on out-of-bag classification error rate (OOB) and variable important spectrum (IS). Our results support that RF(OOB) had better performance than MARS and RF(IS) in detecting important variables. This study demonstrates that TRM(OOB) , which is RF(OOB) plus MARS, has combined the strengths of RF and MARS in identifying SNP-SNP interactions in a scenario of 100 candidate SNPs. TRM(OOB) had greater true positive rate and lower false positive rate compared with MARS, particularly for searching interactions with a strong association with the outcome. Therefore, the use of TRM(OOB) is favored for exploring SNP-SNP interactions in a large-scale genetic variation study.

A novel survival multifactor dimensionality reduction method for detecting gene–gene interactions with application to bladder cancer prognosis

The widespread use of high-throughput methods of single nucleotide polymorphism (SNP) genotyping has created a number of computational and statistical challenges. The problem of identifying SNP-SNP interactions in case-control studies has been studied extensively, and a number of new techniques have been developed. Little progress has been made, however, in the analysis of SNP-SNP interactions in relation to time-to-event data, such as patient survival time or time to cancer relapse. We present an extension of the two class multifactor dimensionality reduction (MDR) algorithm that enables detection and characterization of epistatic SNP-SNP interactions in the context of survival analysis. The proposed Survival MDR (Surv-MDR) method handles survival data by modifying MDR's constructive induction algorithm to use the log-rank test. Surv-MDR replaces balanced accuracy with log-rank test statistics as the score to determine the best models. We simulated datasets with a survival outcome related to two loci in the absence of any marginal effects. We compared Surv-MDR with Cox-regression for their ability to identify the true predictive loci in these simulated data. We also used this simulation to construct the empirical distribution of Surv-MDR's testing score. We then applied Surv-MDR to genetic data from a population-based epidemiologic study to find prognostic markers of survival time following a bladder cancer diagnosis. We identified several two-loci SNP combinations that have strong associations with patients' survival outcome. Surv-MDR is capable of detecting interaction models with weak main effects. These epistatic models tend to be dropped by traditional Cox regression approaches to evaluating interactions. With improved efficiency to handle genome wide datasets, Surv-MDR will play an important role in a research strategy that embraces the complexity of the genotype-phenotype mapping relationship since epistatic interactions are an important component of the genetic basis of disease.