Elucidation of Gene Interaction Networks Through Time-Lagged Correlation Analysis of Transcriptional Data

Abstract

The photosynthetic cyanobacterium Synechocystis sp. strain PCC 6803 uses a complex genetic program to control its physiological response to alternating light conditions. To study this regulatory program time-series experiments were conducted by exposing Synechocystis sp. to serial perturbations in light intensity. In each experiment whole-genome DNA microarrays were used to monitor gene transcription in 20-min intervals over 8-and 16-h periods. The data was analyzed using time-lagged correlation analysis, which identifies genetic interaction networks by constructing correlations between time-shifted transcription profiles with different levels of statistical confidence. These networks allow inference of putative cause-effect relationships among the organism's genes. Using light intensity as our initial input signal, we identified six groups of genes whose time-lagged profiles possessed significant correlation, or anti-correlation, with the light intensity. We expanded this network by using the average profile from each group of genes as a seed, and searching for other genes whose time-lagged profiles possessed significant correlation, or anti-correlation, with the group's average profile. The final network comprised 50 different groups containing 259 genes. Several of these gene groups possess known light-stimulated gene clusters, such as Synechocystis sp. photosystems I and II and carbon dioxide fixation pathways, while others represent novel findings in this work.The DNA microarray has become an established tool for the parallel monitoring of gene expression profiles. Most common experimental design strategies observe static gene expression differences between conditions, such as disease versus nondisease case comparisons. While such experiments generate information for diagnostic applications, they are not well suited for uncovering the roles of these genes in the larger context of cellular regulation.Dynamic transcriptional data allow the formation of gene clusters with similar temporal expression profiles. The various forms of clustering Alter et al. 2000;Holter et al. 2000) employed to date have produced valuable information, including potential gene relationships and the identity of transcription factor binding motifs. These methods, however, are limited in their ability to infer causality or directional relationships between genes. The results of clustering algorithms often yield relations such as "gene A is a good predictor of gene B," which is an equivalent statement to "gene B is a good predictor of gene A." Neither Bayesian networks (Friedman et al. 2000), nor information theory-based approaches (Somogyi and Fuhrman 1997) have made use of the sequential nature of time-series data in current applications. Conversely, when enough time points are available to prevent over fitting the data and find statistically significant correlations, a discovery method to uncover potential causal relationships among genes may be attempted. Directionality can be added to these probabilistic networks by determining the...

Characterization of early autophagy signaling by quantitative phosphoproteomics

“…For this we applied the time-lagged correlation 35 strategy outlined in Figure S5 that uses a correlation coefficient to infer protein-protein interactions by determining the relationship between the temporal dynamics of 2 phosphosites. We applied this to all phosphosites on proteins in the MTOR signaling pathway and by combining the quantitative information after both treatments obtained a regulatory network for potential interactions between MTOR signaling proteins containing 15 protein-protein interactions, based on data from 39 sites (Table S4).…”

Algorithms for modeling global and context-specific functional relationship networks

“…gene A is regulated by gene B after some delay, which makes the two genes correlated with each other with a time lag. This relationship is one type of the stage-specific relationship and can be inferred by algorithms taking the delays into consideration [101,102].…”

Reverse-engineering biological networks from large data sets

“…This imposes uniformity not only on the question of network inference itself, but also on the obstacles and algorithmic approaches that underlie reconstruction efforts across multiple biological domains. Inference methods designed for one system type ( [63], [64], and [65]) can often be adapted to accommodate others ( [66][67][68], [32,69,70], and [39,71], respectively). Moreover, morally equivalent methods have been developed in nominally unrelated fields [72] -or else borrowed explicitly from established disciplines, such as systems identification techniques migrating to network biology from engineering [73,74]).…”