Medicago truncatula gene regulatory network prediction server

摘要

Medicago truncatula, a model legume species, plays an important role for study of gene regulatory mechanisms that are specific to legume biology, such as the process of symbiotic nitrogen fixation which is an unique function of legumes. The aim of this project is to provide the legume community a web-based computational service for predicting and testing hypotheses on Medicago gene regulatory networks (GRNs) that govern various biological processes of legume crops. To achieve this goal, the Medicago microarray data sets have been collected from our internal research groups, collaborators and public databases. Then, three GRNs related to nodule development, seed development and nod factor responses were predicted and populated for public access. Besides these statistic data, the web server provides a user-friendly interface to allow users' access to a personalized GRN prediction pipeline which is backed with a job scheduler. Users would be able to customize gene list, as well as the different algorithms and parameters to be used for GRN prediction. Current implemented algorithms include co-expression, Graphical Gaussian Models, Bayesian networks, Algorithm for the Reconstruction of Accurate Cellular Networks (ARACNE) and Low-order constraint based PC algorithm (LPC) which was our published novel algorithm. In the near future, a web-based visualization and interactive interpretation component will be implemented for users' navigation and annotation of GRNs intuitively.