The Use of the MiDA Software for Establishing Molecular Genetic Markers of the Risk of Developing Infectious Disease Complications

Introduction. The search for specific molecular and genetic markers of the risk of developing infectious disease complications is a current area of research in modern medical and biological science. Materials and methods. In order to solve this issue, we developed a MiDA software that implements an integrated approach allowing for selection of potential markers on the basis of indicators of expression fold change of a number of genes in the comparison groups and the feature importance for classification, i.e. the assignment of samples to the analyzed groups. Results. Using the MiDA software, we searched for molecular and genetic markers of the risk of developing severe dengue fever and chronic brucellosis. As a result of the study, the HSPA6 gene was proposed as a risk marker for the dengue complication. HSPA6 expression was reduced in the peripheral blood samples of severe dengue cases. Markers of chronic brucellosis included a decrease in the expression of miRNA hsa-miR-198 and hsa-miR-501-3p, as well as an increase in the expression of miRNA hsa-miR-618 in CD4+ T-lymphocytes. Conclusion. We demonstrated the possibility of applying the MiDA software to the analysis of big data obtained using modern techniques (sequencing, biochips, etc.). It is possible to expand the scope of the software application in order to analyze the expression of genes, transcripts and proteins in diseases of various origins, to determine molecular mechanisms of the pathological process, to search for diagnostic and prognostic markers of the disease, as well as potential targets for the development of specific therapies.

анализ данных, биочип, большие данные, бруцеллез, лихорадка денге, машинное обучение, программа для ЭВМ, риск осложнений, R, data analysis, biochip, big data, brucellosis, dengue fever, machine learning, software, risk of complications, R

1. Железникова Г.Ф. Инфекция и иммунитет: стратегии обеих сторон. Медицинская Иммунология. 2006. Т. 8. № 5-6. С. 597-614. II. Логвиненко О.В., Ракитина Е.Л., Пономаренко Д.Г. и др. Особенности иммунологических показателей крови у больных различными формами бруцеллеза. Инфекция и иммунитет. 2013. Т. 3. № 3. С. 275-278.

2. Cao Z, Wang Y, Sun Y, et al. Effective and stable feature selection method based on filter for gene signature identification in paired microarray data [abstract]. 2013 IEEE International Conference on Bioinformatics and Biomedicine. 2013. P. 189-192.

3. McCarthy DJ, Smyth GK. Testing significance relative to a fold-change threshold is a TREAT. Bioinformatics. 2009; 25(6):765-71. DOI:10.1093/bioinformatics/btp053

4. Adewale AJ, Dinu I, Yasui Y. Boosting for correlated binary classification. J Comput Graph Stat. 2010; 19(1):140-153.

5. Friedman J, Hastie T, Tibshirani R. Additive logistic regression: a statistical view of boosting (With discussion and a rejoinder by the authors). Ann Statist. 2000; 28(2):337-407.

6. Mieth B, Kloft M, Rodriguez JA, et al. Combining multiple hypothesis testing with machine learning increases the statistical power of genome-wide association studies. Sci Rep. 2016; 6:36671. DOI:10.1038/srep36671

7. Huggins CE, Domenighetti AA, Ritchie ME, et al. Functional and metabolic remodelling in GLUT4-deficient hearts confers hyper-responsiveness to substrate intervention. J Mol Cell Cardiol. 2008; 44(2):270-80. DOI:10.1016/ j.yjmcc.2007.11.020

8. Jeffery IB, Higgins DG, Culhane AC. Comparison and evaluation of methods for generating differentially expressed gene lists from microarray data. BMC Bioinformatics. 2006; 7:359. DOI:10.1186/1471-2105-7-359

9. Pirooznia M, Yang JY, Yang MQ, et al. A comparative study of different machine learning methods on microarray gene expression data. BMC Genomics. 2008; 9(Suppl 1):S13. DOI:10.1186/1471-2164-9-S1-S13

10. Kondrikov D, Fulton D, Dong Z, et al. Heat shock protein 70 prevents hyperoxia-induced disruption of lung endothelial barrier via caspase-dependent and AIF-dependent pathways. PLoS One. 2015; 10(6):e0129343. DOI:10.1371/journal. pone.0129343

11. Wei D, Miao Y, Yu L, et al. Downregulation of microRNA-198 suppresses cell proliferation and invasion in retinoblastoma by directly targeting PTEN. Mol Med Rep. 2018; 18(1):595-602. DOI: 10.3892/mmr.2018.8979

12. Yi L, Yuan Y. MicroRNA-618 modulates cell growth via targeting PI3K/Akt pathway in human thyroid carcinomas. Indian J Cancer. 2015; 52(Suppl 3):E186-189. DOI: 10.4103/0019-509X.186577

13. Zhang CG, Yang F, Li YH, et al. miR-501-3p sensitizes glioma cells to cisplatin by targeting MYCN. Mol Med Rep. 2018; 18(5):4747-4752. DOI: 10.3892/mmr.2018.9458

14. Song L, Ara T, Wu HW, et al. Oncogene MYCN regulates localization of NKT cells to the site of disease in neuroblastoma. J Clin Invest. 2007; 117(9):2702-12. DOI:10.1172/JCI30751