Human Cancer Classification and Prediction Based on Gene Profiling

Abstract

Cancer is one of the larger families of diseases; it is a collection of 100s of diseases that involve abnormal cell to grow and spread to the other parts of the body which leads to worldwide death of the human being. There are different types of cancer, so for those we need to identify and classify all those different types. In the field of bioinformatics, the main important thing is cancer diagnosis, so we need to do it by selecting the subset of feature gene. In cancer diagnosis, the greatest significance is classifying the different types of tumors. By providing an accurate prediction for various types of tumors, we can provide a better treatment for cancer patients and also it reduces the toxicity on patients. The main important thing in this paper is to differentiate between cancer subtypes by creating different methodologies. This paper explains different methodologies, based on gene profiling for the classification and prediction for different types of human cancer. The proposed methodology in this paper is a combination of symmetrical uncertainty (SU) and genetic algorithm (GA).