The Performance Evaluation of The Random Forest Algorithm for A Gene Selection in Identifying Genes Associated with Resectable Pancreatic Cancer in Microarray Dataset: A Retrospective Study

OBJECTIVE: In microarray datasets, hundreds and thousands of genes are measured in a small number of samples, and sometimes due to problems that occur during the experiment, the expression value of some genes is recorded as missing. It is a difficult task to determine the genes that cause disease or cancer from a large number of genes. This study aimed to find effective genes in pancreatic cancer (PC). First, the K-nearest neighbor (KNN) imputation method was used to solve the problem of missing values (MVs) of gene expression. Then, the random forest algorithm was used to identify the genes associated with PC. MATERIALS AND METHODS: In this retrospective study, 24 samples from the GSE14245 dataset were examined. Twelve samples were from patients with PC, and 12 samples were from healthy control. After preprocessing and applying the fold-change technique, 29482 genes were used. We used the KNN imputation method to impute when a particular gene had MVs. Then, the genes most strongly associated with PC were selected using the random forest algorithm. We classified the dataset using support vector machine (SVM) and naïve bayes (NB) classifiers, and F-score and Jaccard indices were reported. RESULTS: Out of the 29482 genes, 1185 genes with fold-changes greater than 3 were selected. After selecting the most associated genes, 21 genes with the most important value were identified. S100P and GPX3 had the highest and lowest importance values, respectively. The F-score and Jaccard value of the SVM and NB classifiers were 95.5, 93, 92, and 92 percent, respectively. CONCLUSION: This study is based on the application of the fold change technique, imputation method, and random forest algorithm and could find the most associated genes that were not identified in many studies. We therefore suggest researchers use the random forest algorithm to detect the related genes within the disease of interest.