MOTIVATION: DNA microarrays are routinely applied to study diseased or drug-treated cell populations. A critical challenge is distinguishing the genes directly affected by these perturbations from the hundreds of genes that are indirectly affected. Here, we developed a sparse simultaneous equation model (SSEM) of mRNA expression data and applied Lasso regression to estimate the model parameters, thus constructing a network model of gene interaction effects. This inferred network model was then used to filter data from a given experimental condition of interest and predict the genes directly targeted by that perturbation. RESULTS: Our proposed SSEM-Lasso method demonstrated substantial improvement in sensitivity compared with other tested methods for predicting the targets of perturbations in both simulated datasets and microarray compendia. In simulated data, for two different network types, and over a wide range of signal-to-noise ratios, our algorithm demonstrated a 167% increase in sensitivity on average for the top 100 ranked genes, compared with the next best method. Our method also performed well in identifying targets of genetic perturbations in microarray compendia, with up to a 24% improvement in sensitivity on average for the top 100 ranked genes. The overall performance of our network-filtering method shows promise for identifying the direct targets of genetic dysregulation in cancer and disease from expression profiles. AVAILABILITY: Microarray data are available at the Many Microbe Microarrays Database (M3D, http://m3d.bu.edu). Algorithm scripts are available at the Gardner Lab website (http://gardnerlab.bu.edu/SSEMLasso).

Growing concerns about global climate change, biodiversity maintenance, natural resources conservation, and long-term ecosystem sustainability have been responsible for the transformation of traditional single resource management approaches into integrated ecosystem management models. Eco-regions are large ecosystems of regional extent that contain smaller ecosystems of similar response potential and resource production capabilities. They can be used as a geographical framework for organizing and reporting resource information, setting bioecological recovery criteria, extrapolating site-level management, and monitoring global change. The objective of this research is to develop a quantitative, multivariate regionalization model that is capable of delineating eco-regions at multiple levels from remotely sensed information and other environmental and natural resources spatial data. The Spatial Pattern Analysis Model developed in this study uses a region-growing algorithm to generate spatially contiguous regions from primitive polygonal land units. The algorithm merges the most similar pair of neighbouring units at each iteration, based on satisfying certain similarity criteria until all units are grouped into one. This model was utilized to develop an eco-region map of Nebraska with three hierarchical levels. In the mapping process, the STATSGO data set was used to build the primitive map units. Environmental parameters included in the model were multi-temporal AVHRR data, soil rooting depth, organic matter content, available water capacity, and long-term annual averages of water balance and growing degree day totals. Development of the model provides a new and useful approach to eco-region mapping for resource managers and researchers. The method is automated and efficient, reduces the judgement biases and uncertainty of manual analyses, and can be replicated for other regions or for the regionalization of other themes.