TRANS-MNET (transcriptional module Network) performs State Space Model to time-course microarray data. State Space Model is a statistical model for analyzing time-series data and state space model implemented in TRANS-MNET is optimized for microarray data. The typical microarray time-course data is high dimensional, but has few time-points, TRANS-MNET can use replicated time-courses includes biological and technical replicates. Also, parameter constraint imposed in TRANS-MNET yields the first-order autoregressive representation of state space models that can be viewed as a parsimonius parameterization of vector AR(1). The permutation test can be applied for finding significance of its AR coefficient and this achives gene regulatory networks.
- R package
:: MORE INFORMATION
R.Yoshida, T.Higuchi S.Imoto,
Estimating time-dependent gene networks from time series microarray data by dynamic linear models with Markov swotching,
Proc.4th Computational Systems Bioinformatics (CSB2005: Refereed Conference), 289-298, 2005.