Using the Precision Lasso for gene selection in diffuse large B cell lymphoma cancer

Abstract

Abstract Background Gene selection from gene expression profiles is the appropriate tool for diagnosing and predicting cancers. The aim of this study is to perform a Precision Lasso regression model on gene expression of diffuse large B cell lymphoma patients and to find marker genes related to DLBCL. Methods In the present case–control study, the dataset included 180 gene expressions from 14 healthy individuals and 17 DLBCL patients. The marker genes were selected by fitting Ridge, Lasso, Elastic Net, and Precision Lasso regression models. Results Based on our findings, the Precision Lasso, the Ridge, the Elastic Net, and the Lasso models choose the most marker genes, respectively. In addition, the top 20 genes are based on models compared with the results of clinical studies. The Precision Lasso and the Ridge models selected the most common genes with the clinical results, respectively. Conclusions The performance of the Precision Lasso model in selecting related genes could be considered more acceptable rather than other models.

Keywords

• Gene expression
• High-dimensional
• Precision Lasso
• Lymphoma cancer