Dataset Information

Network inference performance complexity: a consequence of topological, experimental and algorithmic determinants.

ABSTRACT: MOTIVATION:Network inference algorithms aim to uncover key regulatory interactions governing cellular decision-making, disease progression and therapeutic interventions. Having an accurate blueprint of this regulation is essential for understanding and controlling cell behavior. However, the utility and impact of these approaches are limited because the ways in which various factors shape inference outcomes remain largely unknown. RESULTS:We identify and systematically evaluate determinants of performance-including network properties, experimental design choices and data processing-by developing new metrics that quantify confidence across algorithms in comparable terms. We conducted a multifactorial analysis that demonstrates how stimulus target, regulatory kinetics, induction and resolution dynamics, and noise differentially impact widely used algorithms in significant and previously unrecognized ways. The results show how even if high-quality data are paired with high-performing algorithms, inferred models are sometimes susceptible to giving misleading conclusions. Lastly, we validate these findings and the utility of the confidence metrics using realistic in silico gene regulatory networks. This new characterization approach provides a way to more rigorously interpret how algorithms infer regulation from biological datasets. AVAILABILITY AND IMPLEMENTATION:Code is available at http://github.com/bagherilab/networkinference/. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.

SUBMITTER: Muldoon JJ

PROVIDER: S-EPMC6748731 | biostudies-literature | 2019 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Network inference performance complexity: a consequence of topological, experimental and algorithmic determinants.

Muldoon Joseph J JJ Yu Jessica S JS Fassia Mohammad-Kasim MK Bagheri Neda N

Bioinformatics (Oxford, England) 20190901 18

<h4>Motivation</h4>Network inference algorithms aim to uncover key regulatory interactions governing cellular decision-making, disease progression and therapeutic interventions. Having an accurate blueprint of this regulation is essential for understanding and controlling cell behavior. However, the utility and impact of these approaches are limited because the ways in which various factors shape inference outcomes remain largely unknown.<h4>Results</h4>We identify and systematically evaluate de ...[more]

PMID: 30932143

Similar Datasets

Project description:A common way of illustrating phylogeographic results is through the use of haplotype networks. While these networks help to visualize relationships between individuals, populations, and species, evolutionary studies often only quantitatively analyze genetic diversity among haplotypes and ignore other network properties. Here, we present a new metric, haplotype network branch diversity (HBd), as an easy way to quantifiably compare haplotype network complexity. Our metric builds off the logic of combining genetic and topological diversity to estimate complexity previously used by the published metric haplotype network diversity (HNd). However, unlike HNd which uses a combination of network features to produce complexity values that cannot be defined in probabilistic terms, thereby obscuring the values' implication for a sampled population, HBd uses frequencies of haplotype classes to incorporate topological information of networks, keeping the focus on the population and providing easy-to-interpret probabilistic values for randomly sampled individuals. The goal of this study is to introduce this more intuitive metric and provide an R script that allows researchers to calculate diversity and complexity indices from haplotype networks. A group of datasets, generated manually (model dataset) and based on published data (empirical dataset), were used to illustrate the behavior of HBd and both of its terms, haplotype diversity, and a new index called branch diversity. Results followed a predicted trend in both model and empirical datasets, from low metric values in simple networks to high values in complex networks. In short, the new combined metric joins genetic and topological diversity of haplotype networks, into a single complexity value. Based on our analysis, we recommend the use of HBd, as it makes direct comparisons of network complexity straightforward and provides probabilistic values that can readily discriminate situations that are difficult to resolve with available metrics.

Dataset Information

Network inference performance complexity: a consequence of topological, experimental and algorithmic determinants.

Publications

Network inference performance complexity: a consequence of topological, experimental and algorithmic determinants.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets