TY - GEN
T1 - Using Chou's amphiphilic Pseudo-Amino Acid Composition and Extreme Learning Machine for prediction of Protein-protein interactions
AU - Huang, Qiao Ying
AU - You, Zhu Hong
AU - Li, Shuai
AU - Zhu, Zexuan
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2014/9/3
Y1 - 2014/9/3
N2 - Protein-protein interactions (PPIs) play crucial roles in the execution of various cellular processes. Almost every cellular process relies on transient or permanent physical bindings of proteins. Unfortunately, the experimental methods for identifying PPIs are both time-consuming and expensive. Therefore, it is important to develop computational approaches for predicting PPIs. In this study, a novel approach is presented to predict PPIs using only the information of protein sequences. This method is developed based on learning algorithm-Extreme Learning Machine (ELM) combined with the concept of Chous Pseudo-Amino Acid Composition (PseAAC) composition. PseAAC is a combination of a set of discrete sequence correlation factors and the 20 components of the conventional amino acid composition, so this method can observe a remarkable improvement in prediction quality. ELM classifier is selected as prediction engine, which is a kind of accurate and fast-learning innovative classification method based on the random generation of the input-to-hidden-units weights followed by the resolution of the linear equations to obtain the hidden-to-output weights. When performed on the PPIs data of Saccharomyces cerevisiae, the proposed method achieved 79.66% prediction accuracy with 79.16% sensitivity at the precision of 79.96%. Extensive experiments are performed to compare our method with state-of-the-art techniques Support Vector Machine (SVM). Achieved results show that the proposed approach is very promising for predicting PPIs, and it can be a helpful supplement for PPIs prediction.
AB - Protein-protein interactions (PPIs) play crucial roles in the execution of various cellular processes. Almost every cellular process relies on transient or permanent physical bindings of proteins. Unfortunately, the experimental methods for identifying PPIs are both time-consuming and expensive. Therefore, it is important to develop computational approaches for predicting PPIs. In this study, a novel approach is presented to predict PPIs using only the information of protein sequences. This method is developed based on learning algorithm-Extreme Learning Machine (ELM) combined with the concept of Chous Pseudo-Amino Acid Composition (PseAAC) composition. PseAAC is a combination of a set of discrete sequence correlation factors and the 20 components of the conventional amino acid composition, so this method can observe a remarkable improvement in prediction quality. ELM classifier is selected as prediction engine, which is a kind of accurate and fast-learning innovative classification method based on the random generation of the input-to-hidden-units weights followed by the resolution of the linear equations to obtain the hidden-to-output weights. When performed on the PPIs data of Saccharomyces cerevisiae, the proposed method achieved 79.66% prediction accuracy with 79.16% sensitivity at the precision of 79.96%. Extensive experiments are performed to compare our method with state-of-the-art techniques Support Vector Machine (SVM). Achieved results show that the proposed approach is very promising for predicting PPIs, and it can be a helpful supplement for PPIs prediction.
KW - Extreme Learning Machine(ELM)
KW - Protein-protein Interactions
KW - Pseudo-amino Acid Composition
KW - Support Vector Machine (SVM)
UR - http://www.scopus.com/inward/record.url?scp=84908471908&partnerID=8YFLogxK
U2 - 10.1109/IJCNN.2014.6889476
DO - 10.1109/IJCNN.2014.6889476
M3 - 会议稿件
AN - SCOPUS:84908471908
T3 - Proceedings of the International Joint Conference on Neural Networks
SP - 2952
EP - 2956
BT - Proceedings of the International Joint Conference on Neural Networks
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2014 International Joint Conference on Neural Networks, IJCNN 2014
Y2 - 6 July 2014 through 11 July 2014
ER -