Research Article 
Autoregressive-Model Based Dynamic Fuzzy Clustering for Time-Course Gene Expression Data
Xu Hong-Lin, Liu Yu-Hong and Wang Shi-Tong
Biotechnology, 2008, 7(1), 59-65.
Abstract
In this study, a novel algorithm called Dynamic Fuzzy Clustering (DFC) is proposed for clustering time-course gene expression data. The proposed method combines Autoregressive (AR) model and conventional Fuzzy Clustering Algorithm (FCM). Under this approach, a time-course gene expression data can be analyzed as a set of dynamic time series with AR model in order to utilize the important dynamic information more efficiently and the forecast process in AR model can be adjusted using the corresponding fuzzy membership such that better clustering results can be obtained. Experiments performed on a synthetic and two real-world time-course gene expression datasets also indicates that this proposed approach can be more effective than some other conventional clustering algorithms such as FCM and simple dynamic model-based clustering algorithm.
ASCI-ID: 11-356