TY - GEN
T1 - A Novel Method for Identifying Alzheimer's Disease-related Proteins
AU - Hu, Yang
AU - Zhang, Jun
AU - Zhao, Tianyi
AU - Cheng, Liang
AU - Zang, Tianyi
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2019/1/21
Y1 - 2019/1/21
N2 - Alzheimer's disease (AD) is a type of neurodegenerative disease that has become the fourth 'health killer' for the elderly after cardiovascular, malignant, and stroke. Identifying AD-related proteins can effectively help diagnose diseases in advance, discover new drug targets, and gain a deeper understanding of the pathogenesis of disease. Due to the high cost of biological experiments, more and more researchers introduced advanced algorithms into this field. Based on the hypothesis of 'similar diseases shares similar related proteins in this paper, diseases which are similar to AD were found by five similarity calculation methods. Then, related proteins of each disease were obtained by public data set. Through these proteins, features are extracted. Then, these features are mapped to disease similarity by Logistic Regression (LR). To avoid curse of dimensionality, random features are selected each time and hundreds of models were built. For each model, AD-related proteins could be obtained by Gradient Descent method. Finally, we integrated all the models and get all the proteins related to AD by weight counting. Finally, we did three case studies to prove novel proteins found by us are reliable.
AB - Alzheimer's disease (AD) is a type of neurodegenerative disease that has become the fourth 'health killer' for the elderly after cardiovascular, malignant, and stroke. Identifying AD-related proteins can effectively help diagnose diseases in advance, discover new drug targets, and gain a deeper understanding of the pathogenesis of disease. Due to the high cost of biological experiments, more and more researchers introduced advanced algorithms into this field. Based on the hypothesis of 'similar diseases shares similar related proteins in this paper, diseases which are similar to AD were found by five similarity calculation methods. Then, related proteins of each disease were obtained by public data set. Through these proteins, features are extracted. Then, these features are mapped to disease similarity by Logistic Regression (LR). To avoid curse of dimensionality, random features are selected each time and hundreds of models were built. For each model, AD-related proteins could be obtained by Gradient Descent method. Finally, we integrated all the models and get all the proteins related to AD by weight counting. Finally, we did three case studies to prove novel proteins found by us are reliable.
KW - Alzheimer's disease
KW - gradient Descent
KW - logistic Regression
KW - proteins
KW - similarity of diseases
UR - https://www.scopus.com/pages/publications/85062572780
U2 - 10.1109/BIBM.2018.8621492
DO - 10.1109/BIBM.2018.8621492
M3 - 会议稿件
AN - SCOPUS:85062572780
T3 - Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
SP - 1256
EP - 1261
BT - Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
A2 - Schmidt, Harald
A2 - Griol, David
A2 - Wang, Haiying
A2 - Baumbach, Jan
A2 - Zheng, Huiru
A2 - Callejas, Zoraida
A2 - Hu, Xiaohua
A2 - Dickerson, Julie
A2 - Zhang, Le
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
Y2 - 3 December 2018 through 6 December 2018
ER -