Tracking the states of a nonlinear system in the weight-space of a feed-forward neural network

Nonlinear, non-stationary signals are commonly found in a variety of disciplines such as biology, medicine, geology and financial modeling. The complexity (e.g. nonlinearity and non-stationarity) of such signals and their low signal to noise ratios often make it a challenging task to use them in critical applications. In this paper we propose a new neural network based technique to address those problems. We show that a feed forward, multi-layered neural network can conveniently capture the states of a nonlinear system in its connection weight-space, after a process of supervised training. The performance of the proposed method is investigated via computer simulations.

Multilevel modelling for inference of genetic regulatory networks

Time-course experiments with microarrays are often used to study dynamic biological systems and genetic regulatory networks (GRNs) that model how genes influence each other in cell-level development of organisms. The inference for GRNs provides important insights into the fundamental biological processes such as growth and is useful in disease diagnosis and genomic drug design. Due to the experimental design, multilevel data hierarchies are often present in time-course gene expression data. Most existing methods, however, ignore the dependency of the expression measurements over time and the correlation among gene expression profiles. Such independence assumptions violate regulatory interactions and can result in overlooking certain important subject effects and lead to spurious inference for regulatory networks or mechanisms. In this paper, a multilevel mixed-effects model is adopted to incorporate data hierarchies in the analysis of time-course data, where temporal and subject effects are both assumed to be random. The method starts with the clustering of genes by fitting the mixture model within the multilevel random-effects model framework using the expectation-maximization (EM) algorithm. The network of regulatory interactions is then determined by searching for regulatory control elements (activators and inhibitors) shared by the clusters of co-expressed genes, based on a time-lagged correlation coefficients measurement. The method is applied to two real time-course datasets from the budding yeast (Saccharomyces cerevisiae) genome. It is shown that the proposed method provides clusters of cell-cycle regulated genes that are supported by existing gene function annotations, and hence enables inference on regulatory interactions for the genetic network.

Left-right information flow in the brain during EEG arousals

In disorders such as sleep apnea, sleep is fragmented with frequent EEG-arousal (EEGA) as determined via changes in the sleep-electroencephalogram. EEGA is a poorly understood, complicated phenomenon which is critically important in studying the mysteries of sleep. In this paper we study the information flow between the left and right hemispheres of the brain during the EEGA as manifested through inter-hemispheric asynchrony (IHA) of the surface EEG. EEG data (using electrodes A1/C4 and A2/C3 of international 10-20 system) was collected from 5 subjects undergoing routine polysomnography (PSG). Spectral correlation coefficient (R) was computed between EEG data from two hemispheres for delta-delta(0.5-4 Hz), theta-thetas(4.1-8 Hz), alpha-alpha(8.1-12 Hz) & beta-beta(12.1-25 Hz) frequency bands, during EEGA events. EEGA were graded in 3 levels as (i) micro arousals (3-6 s), (ii) short arousals (6.1-10 s), & (iii) long arousals (10.1-15 s). Our results revealed that in beta band, IHA increases above the baseline after the onset of EEGA and returns to the baseline after the conclusion of event. Results indicated that the duration of EEGA events has a direct influence on the onset of IHA. The latency (L) between the onset of arousals and IHA were found to be L=2plusmn0.5 s (for micro arousals), 4plusmn2.2 s (short arousals) and 6.5plusmn3.6 s (long arousals)