Development of bioinformatics software using data mining method in artificial intelligence: An application with citrulline

Artificial intelligence has started to gain importance especially in the field of health today. Data mining is an interdisciplinary field that combines artificial intelligence, computer science, machine learning, data visualization, database management, algorithms and statistics. The main subject of data mining is to select the needed daha among the all data, and then to process them and create meaningful connections. In this study, it is aimed to create an interface using R Software with data mining of citrulline, which is an important protein/aminoacid in human metabolism, whose biosynthesis and metabolic pathway has not been fully elucidated yet, with a statistics-based approach. Citrulline-related gene data for this study were obtained from the GenBank Overview-NCBI-GEO database. Bayesian Classification Algorithm was used on the data and bioinformatic analysis was performed. The effects of citrulline-related genes on the protein were numerically shown and correlated with the analyzes. After the association, String, Reactome Pathway Database, Wiki Pathway, KEGG and David Gen Pathway databases were used to determine possible metabolic pathways. Considering the pathways in which citrulline-related genes are involved; While CPS1 was the most involved gene in the biosynthesis pathway, ASS1 was the gene most involved in the metabolic pathway. In the metabolic pathway analysis, it has been determined that the pathway in which citrulline is most involved and effective is the urea cycle with a rate of 30%. In this context, determined as an example of clinical picture, Cluster analysis (heat-map analysis) of 100 citrulline-related genes in the urea cycle was performed through data from clinical patients with kidney transplantation and tissue damage. Results were interpreted by PlotMD, VolcanoPlot, BoxPlot, QQ-Plot analysis.