论文部分内容阅读
Comparison of current knowledge on plant transcriptional regulatory networks (TRNs)is sparse, especially in comparison with the state of art for E.coil and Yeast.Moreover most of the known transcription factor [TF]-target interactions concern developmental processes such as seed (Ge et al.2010), and flower development (Wellmer and Riechmann 2010) or meristematic growth (Hamant and Pautot 2010).In our study we have addressed the largely unknown field of TRNs responsible for gene expression regulation in response to the changes of most common and tightly coupled environmental parameters: temperature and light intensity.Here, we present analysis of the early transcriptional responses of Arabidopsis thaliana plants exposed to different temperatures (4 ℃, 21 ℃ and 32 ℃) and light intensities (darkness, 75μE, 150 μE and 400 μE), including selected combinations.We define specificity of differentially expressed genes towards particular treatments and particular light or temperature regimes and use a temporal logic based approach to extract robust temporal patterns of consecutive transcriptional changes and to define the space of possible TF-target interactions.The resulting graph representation of this quantitative time series data consists of genes being nodes connected by directed edges of activation or repression type.Integrating the criteria of mutual information between gene expression, gene co-expression in independent experiments and composition of cis-elements in the promoter regions we extract the most significant, thus most promising, network components representing hypothetical interactions between elements of Arabidopsis thaliana early response to changes in light and temperature.In order to confirm created hypotheses we performed a range of follow up experiments, including generation and qRT-PCR analysis of inducible TFs overexpression lines, stress survival tests of chosen TF knockout lines.This study represents a classical systems-biology approach towards the creation of experimentally testable hypotheses from high throughput data, followed by confirmation through targeted genetic approaches.