Deep learning approach for predicting aging-associated genes

• Main Authors: Jheng-yang Sung, 宋政洋
• Other Authors: Jia-Ching Wang
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/gy6pbs

碩士 === 國立中央大學 === 資訊工程學系在職專班 === 107 === Deep learning is the foundation of AI Artificial Intelligence applications. Since the achievement in the field of speech recognition and image recognition, DNN has grown with extremely fast rate in other fields. For biomedicine, the application of deep learning methods, such as cancer detection, bioinformatics analysis, etc., has also been widely used, and aging research has also made significant contributions . In this paper, The genotype tissue from The Genotype-Tissue Expression (GTEx) expresses RNA-seq data for DNA sequencing, with high detection speed, high throughput and wide range of detection. Characteristics, so there is better correctness in detecting gene expression. In this paper, we have tree main directions: 1.Classification and prediction of age groups from normal tissues 2.Compare the results between activation function and loss functions 3.Extracting related gene sets of various tissues by statistical analysis In this paper, we will use machine learning for experiment such like Ridge Regression , Decision Tree , Random Forest, and Support Vector Machine . In order to compare the recognition rates of each method,we also added deep neural network , auto-encoder , and other methods of deep learning.