Microarray target gene extraction with machine learning methods

Processing and analysing large medical data such as microarray, microRNA, dna, rna data is a challenging process. Bioinformatics is a multidisciplinary field of research that enables the processing of such medical data using computational technologies such as machine learning. Studies in the field of bioinformatics, where biology, mathematics and computer science work together, provide support in many areas such as disease diagnosis, preventive treatment, drug development and personalised treatment procedures. Bioinformatics methods are preferred to study the effect of microRNAs (also called miRNA, miR), which are found in various organisms and are thought to be associated with cancer, on genes. Microarray technology is one of the techniques used in the analysis of data such as microRNA, DNA, RNA. Many types of research such as the analysis of gene expression, the extraction of target genes, mutation analysis and the discovery of biomarkers are possible using the microarray method. Different methods can be used to prepare microarray data, and the way in which the data is analysed varies depending on these methods. In this study, biological data prepared using the microarray technique are analysed and a study is conducted to extract target genes. For this purpose, a method using machine learning techniques is developed and applied to different datasets. In addition to two open access datasets, a unique dataset is also used in the study. For 11 experiments belonging to 3 different dataset groups, the process from preparing the data for processing in a computer environment to recognising the target genes is studied in detail. The study includes computerization of raw data, data quality checking, clustering, normalisation, creation of comparison groups and prediction of target genes.