Evolutionary constraints on the complexity of genetic regulatory networks allow predictions of the total number of genetic interactions

Abstract

Genetic regulatory networks (GRNs) have been widely studied, yet there is a lack of understanding with regards to the final size and properties of these networks, mainly due to no network currently being complete. In this study, we analyzed the distribution of GRN structural properties across a large set of distinct prokaryotic organisms and found a set of constrained characteristics such as network density and number of regulators. Our results allowed us to estimate the number of interactions that complete networks would have, a valuable insight that could aid in the daunting task of network curation, prediction, and validation. Using state-of-the-art statistical approaches, we also provided new evidence to settle a previously stated controversy that raised the possibility of complete biological networks being random and therefore attributing the observed scale-free properties to an artifact emerging from the sampling process during network discovery. Furthermore, we identified a set of properties that enabled us to assess the consistency of the connectivity distribution for various GRNs against different alternative statistical distributions. Our results favor the hypothesis that highly connected nodes (hubs) are not a consequence of network incompleteness. Finally, an interaction coverage computed for the GRNs as a proxy for completeness revealed that high-throughput based reconstructions of GRNs could yield biased networks with a low average clustering coefficient, showing that classical targeted discovery of interactions is still needed.

Corynebacterium Glutamicum Regulation beyond Transcription: Organizing Principles and Reconstruction of an Extended Regulatory Network Incorporating Regulations Mediated by Small Rna and Protein-Protein Interactions

“…However, assigning the interaction to the coding gene would be misleading and inflates the number of self-loops in the network even when the sRNAs might be transcribed through its own promoter. As previously discussed, the interaction coverage is a better proxy for network completeness than the genomic coverage [6]. Although the authors provide the σ factors required for the transcription of the sRNAs, we did not include these σ-DNA interactions as they were solely supported by DNA-binding motif computational predictions and we have identified a high number of false positives in the search of binding sites for the σ factors, and purely computational prediction is not considered for the Abasy Atlas networks [1].…”

“…It has been previously debated whether the P(k) of real networks is truly governed by a power-law distribution where a few nodes have most of the interactions [16]. Recently, using several statistic methods we demonstrated that the regulatory networks truly follow a power-law distribution and they would fit other power law-like distributions better than a Poisson distribution regardless of the completeness of the network and that the sole coefficient of determination (R 2 ) is a good proxy to assess the goodness-of-fit of the model [6].…”

“…Namely, E. coli, B. subtilis, and S. coelicolor. Please note that despite the high completeness of the network for M. tuberculosis [1], and the closeness to C. glutamicum in the phylogeny, we did not use this network as a source since it was mainly constructed using only high-throughput technologies without further confirmation with directed experiments causing an unusually lower clustering coefficient of the network (see figure 5 in [6]). Besides, C. glutamicum has been used as a source organism for the inference of regulatory interactions in M. tuberculosis [35].…”

“…Despite being model organisms, their global regulatory networks still incomplete [1] and the scenario is worst for non-model organisms. Even though high-throughput technologies speed up the reconstruction of regulatory networks, it results in biased structural properties of the networks when compared with other reconstructions performed mainly by conventional approaches [6]. Besides, the number of sequenced genomes scales so rapidly, especially for bacteria, that even with high throughput experiments we cannot cope with all of them.…”

Corynebacterium glutamicumregulation beyond transcription: Organizing principles and reconstruction of an extended regulatory network incorporating regulations mediated by small RNA and protein-protein interactions

“…After this update, C. glutamicum moved from the fourth to the second position as the organism with the most complete regulatory network in Abasy Atlas, according to our recently published model of the total number of interactions a complete regulatory network has [6]. We discuss the global structural properties of the three network models in the context of the previous versions for the transcriptional regulatory models and more than 40 other bacterial networks from Abasy Atlas, the most complete collection of experimentally-validated regulatory networks [1].…”