Reconstruction of large-scale regulatory networks based on perturbation graphs and transitive reduction: improved methods and their evaluation

Abstract

BackgroundThe data-driven inference of intracellular networks is one of the key challenges of computational and systems biology. As suggested by recent works, a simple yet effective approach for reconstructing regulatory networks comprises the following two steps. First, the observed effects induced by directed perturbations are collected in a signed and directed perturbation graph (PG). In a second step, Transitive Reduction (TR) is used to identify and eliminate those edges in the PG that can be explained by paths and are therefore likely to reflect indirect effects.ResultsIn this work we introduce novel variants for PG generation and TR, leading to significantly improved performances. The key modifications concern: (i) use of novel statistical criteria for deriving a high-quality PG from experimental data; (ii) the application of local TR which allows only short paths to explain (and remove) a given edge; and (iii) a novel strategy to rank the edges with respect to their confidence. To compare the new methods with existing ones we not only apply them to a recent DREAM network inference challenge but also to a novel and unprecedented synthetic compendium consisting of 30 5000-gene networks simulated with varying biological and measurement error variances resulting in a total of 270 datasets. The benchmarks clearly demonstrate the superior reconstruction performance of the novel PG and TR variants compared to existing approaches. Moreover, the benchmark enabled us to draw some general conclusions. For example, it turns out that local TR restricted to paths with a length of only two is often sufficient or even favorable. We also demonstrate that considering edge weights is highly beneficial for TR whereas consideration of edge signs is of minor importance. We explain these observations from a graph-theoretical perspective and discuss the consequences with respect to a greatly reduced computational demand to conduct TR. Finally, as a realistic application scenario, we use our framework for inferring gene interactions in yeast based on a library of gene expression data measured in mutants with single knockouts of transcription factors. The reconstructed network shows a significant enrichment of known interactions, especially within the 100 most confident (and for experimental validation most relevant) edges.ConclusionsThis paper presents several major achievements. The novel methods introduced herein can be seen as state of the art for inference techniques relying on perturbation graphs and transitive reduction. Another key result of the study is the generation of a new and unprecedented large-scale in silico benchmark dataset accounting for different noise levels and providing a solid basis for unbiased testing of network inference methodologies. Finally, applying our approach to Saccharomyces cerevisiae suggested several new gene interactions with high confidence awaiting experimental validation.

Data- and knowledge-based modeling of gene regulatory networks: an update

“…However, for genome-wide and other large-scale networks the regression-based method LASSO (GENLAB, van Someren et al, 2006[ 161 ]) seems to be best situated if it is well configured and the experimental data and prior knowledge are of sufficient quantity and quality (Marbach et al, 2012[ 99 ]). Boolean and static network modeling should be preferred if the data are mainly steady state gene expression data from KO experiments or for modeling of signaling pathways, respectively (Eduati et al, 2010[ 33 ]; Klamt et al, 2010[ 78 ]; Flassig et al, 2013[ 41 ]; Samaga and Klamt, 2013[ 142 ]; Pinna et al, 2013[ 129 ]; Ryll et al, 2014[ 141 ]; Nakajima and Akutsu, 2014[ 121 ]).…”

TRaCE+: Ensemble inference of gene regulatory networks from transcriptional expression profiles of gene knock-out experiments

“…This simple procedure may not work when G U contains a negative cycle (a directed cycle with an odd number of negative edges), since the cumulative sign of such a cycle alternates depending on how many times one traverses through it. A recent study comparing different ways to obtain the transitive reduction of a signed GRN digraph (with and without cycles) recommended a simple procedure called Local Transitive Reduction (LTR) [ 14 ]. In the following, we have adapted LTR to generate the lower bound signed digraph G L for TRaCE+.…”

“…Here, a positive edge reflects an activation, while a negative edge describes a repression. Several notable network inference algorithms such as TRANSitive reduction for WEighted Signed Digraphs (TRANSWESD) [ 13 ] and Local Transitive Reduction (LTR) [ 14 ] previously considered the inference of GRN digraph with signed (and weighted) edges. However, these algorithms were not designed for inferring an ensemble of GRN structures.…”

Psychological Perturbation Data on Attitudes Towards the Consumption of Meat

“…First, the questionnaire data can aid research that aims to infer causal relations between variables. Since the data contain both observational and experimental data, algorithms like the downward ranking of feed-forward loops (DR-FFL; [4]), the invariant causal prediction (ICP; [5]), or newly created algorithms can be used to investigate to what extent questionnaire items causally influence each other. Second, this data can be used to study the malleability of a person's attitude towards meat consumption, a "hot topic" [2].…”