OCSVM

The software tool OCSVM (One-class Classification Support Vector Machines) is used to identify disease-causing genes, a crucial step in drug design and discovery. Unlike traditional machine learning methods that use known disease genes as positive samples and unknown genes as negative samples to build a binary classification model, OCSVM focuses on creating a model that concentrates on detecting known disease-causing genes to increase sensitivity and precision.

The authors propose using OCSVM to tackle the problem of disease gene identification as a one-class classification problem. The main goal is to identify disease genes while identifying non-disease genes is of less or no significance. They evaluate the performance of their proposed model using a benchmark dataset consisting of gene expression data for Acute Myeloid Leukemia (AML) cancer.