Non-negative Matrix Factorization Algorithm To Deal With The Cancer Gene Expression Data

The occurrence and development of tumor is a complex and multi-stage process. Usually it's because of some gene mutation and abnormal expression or further affecting other genes'expression, which results in the change of protein molecules within cells and produces the tumor differences in the pathology and different classification in the clinical diagnosis. Correctly classify the samples with pathological classification and find the different expression genes is important to the diagnosis and treatment of the tumor.The tumor gene expressing datasets are researched in this paper according to the data characters and biology mechanism. The main work is as follows:Firstly, learning the theory of the non-Negative Matrix Factorization(NMF) method, using it to classify the gastric gene expression data and the classification accuracy reached 98.41%, we also use this method to deal with the colon cancer gene expression data and the correct classification rate is 88.10%. This method gets better results and it has referred important reference to the clinic diagnosis and biomedical research.Secondly, we propose the method of different genes'selection and analyze the function of the genes with obviously different expressions. It is mainly by analyzing the relationship between metagenes and genes, using the Expression Analysis Systematic Explorer(EASE) software and combing some medicine references to discribe the function of the genes with obviously different expressions. The different gene selection method uses the relationship between metagenes and genes smartly to compend the application and enrich the biological significance of the NMF algorithm.