Background: Additional inflammatory responses and subsequent damage—arising from enhance transcriptional activity or forming the more active protein due to existence of polymorphic sites in the pro-inflammatory cytokines gene loci—give rise to myocardial infarction susceptibility. Objectives: The aim of our study was to explore whether two interleukin-13 gene polymorphisms (−1512A/C and +2044G/A) could serve as underpins genetic susceptibility of myocardial infarction. Methods: The Iranian population that belong to the Parsis ethnic group was involved in the present study. A total 250 patients with definite myocardial infarction—meeting hypertension, hypercholesterolemia, hyperglycemia, and coronary artery disease requirements—were recruited from the Shiraz urban hospitals. 250 age- and sex-matched healthy individuals without a history of cardiovascular disease and heart disease related risk factors constituted the control group. PCR-restriction fragment length polymorphism technique applied to genotyping at −1512A/C and +2044G/A loci. Hardy–Weinberg equilibrium test was performed (combined cases and controls). The differences of the genotype frequencies in cases and controls were analyzed using a chi-square test. Logistic regression analysis was performed to assess the association between the genotypes and most important risk factors for myocardial infarction. All statistical analyses were performed in SPSS Version 22.0. p-values below 0.05 were hailed as statistically significant. Results: Deviation from Hardy–Weinberg equilibrium was not significant in the −1512A/C locus. Statistically significant difference between our study groups was found in genotype frequency of the −1512A/C. This variant was found in associated with myocardial infarction risk factors. The +2044G/A polymorphism was not in Hardy–Weinberg equilibrium and no significant difference observed in the distribution of +2044G/A genotype frequency among cases and controls. However, further analysis revealed that this genotype associated with an increased susceptibility to myocardial infarction risk factors. Conclusions: The presence of interleukin-13 −1512A/C and +2044G/A gene polymorphisms underpin myocardial infarction predisposition in the ethnic Parsis of the Iranian population.

The impaired suppressive function of regulatory T cells is well-understood in systemic lupus erythematosus. This is likely due to changes in Foxp₃ expression that are crucial for regulatory T-cell stability and function. There are a few reports on the correlation between the Foxp₃ altered expression level and single-nucleotide polymorphisms within the Foxp₃ locus. Moreover, some studies showed the importance of Foxp₃ expression in the same diseases. Therefore, to explore the possible effects of single-nucleotide polymorphisms, here, we evaluated the association of IVS9+459/rs2280883 (T>C) and −2383/rs3761549 (C>T) Foxp₃ polymorphisms with systemic lupus erythematosus. Moreover, through machine-learning and deep-learning methods, we assessed the connection of the expression level of the gene with the disease. Single-nucleotide polymorphisms of Foxp₃ (IVS9+459/rs2280883 (T>C) and −2383/rs3761549 (C>T)) were, respectively, genotyped using allele-specific PCR and direct sequencing and polymerase chain reaction-restriction fragment length polymorphism, in 199 systemic lupus erythematosus patients and 206 healthy age- and sex-matched controls. The Statistical Package for the Social Sciences version 19 and Fisher’s exact and chi-square tests were used to analyze the data. Moreover, six machine-learning models and two sequential deep-learning models were designed to classify patients from normal people in the E-MTAB-11191 dataset through the expression level of Foxp₃ and its correlated genes. The allele and genotype frequencies of both polymorphisms in question were found to be significantly associated with an increased risk of systemic lupus erythematosus. Furthermore, both of the two single-nucleotide polymorphisms were associated with some systemic-lupus-erythematosus-related risk factors. Three SVM models and the logistic regression model showed an 81% accuracy in classification problems. In addition, the first deep-learning model showed an 83% and 89% accuracy for the training and validation data, respectively, while the second model had an 85% and 79% accuracy for the training and validation datasets. In this study, we are prompted to represent the predisposing loci for systemic lupus erythematosus pathogenesis and strived to provide evidence-based support to the application of machine learning for the identification of systemic lupus erythematosus. It is predicted that the recruiting of machine-learning algorithms with the simultaneous measurement of the applied single nucleotide polymorphisms will increased the diagnostic accuracy of systemic lupus erythematosus, which will be very helpful in providing sufficient predictive value about individual subjects with systemic lupus erythematosus.