Interactive analysis of systems biology molecular expression data

BioNetApp: An interactive visual data analysis platform for molecular expressions

MOTIVATION

Systems biology faces two key challenges when dealing with large amounts of disparate data produced by different experiments: the integration of results across different experiments, and the extraction of meaningful information from the data produced by these experiments. An ongoing challenge is to provide better tools that can mine data patterns that could not have been discovered through simple visualization. Such mining capabilities also need to be coupled with intuitive visualization to portray those findings. We introduce a software toolbox entitled BioNetApp to mine these patterns and visualize them across all experiments.

RESULTS

BioNetApp is an interactive visual data mining software for analyzing high-volume molecular expression data obtained from multiple 'omics experiments. By integrating visualization, statistical methods, and data mining techniques, BioNetApp can perform interactive correlative and comparative analysis along time-course studies of molecular expression data. Correlation analysis provides several visualization features such as Kamada-Kawai, Fruchterman-Reingold Spring embedding network layouts, in addition to single circle, multiple circle and heatmap layouts, whereas comparative analysis presents expression-data distributions across samples, groups, and time points with boxplot display, outlier detection, and data curve fitting. BioNetApp also provides data clustering based on molecular concentrations using Self Organizing Maps (SOM), K-Means, K-Medoids, and Farthest First algorithms.

CONCLUSION

BioNetApp has been utilized in a metabolomics study to investigate the metabolite abundance changes in alcohol induced fatty liver, where pair-wise analyses of metabolome concentration revealed correlation networks and interesting patterns in the metabolomics dataset. This study case demonstrates the effectiveness of the BioNetApp software as an interactive visual analysis tool for molecular expression data in systems biology. The BioNetApp software is freely available under GNU GPL license and can be downloaded (including the case-study data and user-manual) at: https://doi.org/10.5281/zenodo.2563129.

A Systems Biology Approach to Understanding Alcoholic Liver Disease Molecular Mechanism: The Development of Static and Dynamic Models

Alcoholic liver disease (ALD) is a complex disease characterized by damages to the liver and is the consequence of excessive alcohol consumption over years. Since this disease is associated with several pathway failures, pathway reconstruction and network analysis are likely to explicit the molecular basis of the disease. To this aim, in this paper, a network medicine approach was employed to integrate interactome (protein-protein interaction and signaling pathways) and transcriptome data to reconstruct both a static network of ALD and a dynamic model for it. Several data sources were exploited to assemble a set of ALD-associated genes which further was used for network reconstruction. Moreover, a comprehensive literature mining reveals that there are four signaling pathways with crosstalk (TLR4, NF- [Formula: see text]B, MAPK and Apoptosis) which play a major role in ALD. These four pathways were exploited to reconstruct a dynamic model of ALD. The results assure that these two models are consistent with a number of experimental observations. The static network of ALD and its dynamic model are the first models provided for ALD which offer potentially valuable information for researchers in this field.

Identification of transcriptional macromolecular associations in human bone using browser based in silico analysis in a giant correlation matrix

Intracellular signaling is critically dependent on gene regulatory networks comprising physical molecular interactions. Presently, there is a lack of comprehensive databases for most human tissue types to verify such macromolecular interactions. We present a user friendly browser which helps to identify functional macromolecular interactions in human bone as significant correlations at the transcriptional level. The molecular skeletal phenotype has been characterized by transcriptome analysis of iliac crest bone biopsies from 84 postmenopausal women through quantifications of ~23,000 mRNA species. When the signal levels were inter-correlated, an array containing >260 million correlations was generated, thus recognizing the human bone interactome at the RNA level. The matrix correlation and p values were made easily accessible by a freely available online browser. We show that significant correlations within the giant matrix are reproduced in a replica set of 13 male vertebral biopsies. The identified correlations differ somewhat from transcriptional interactions identified in cell culture experiments and transgenic mice, thus demonstrating that care should be taken in extrapolating such results to the in vivo situation in human bone. The current giant matrix and web browser are a valuable tool for easy access to the human bone transcriptome and molecular interactions represented as significant correlations at the RNA-level. The browser and matrix should be a valuable hypothesis generating tool for identification of regulatory mechanisms and serve as a library of transcript relationships in human bone, a relatively inaccessible tissue.

MINER: exploratory analysis of gene interaction networks by machine learning from expression data

Background

The reconstruction of gene regulatory networks from high-throughput "omics" data has become a major goal in the modelling of living systems. Numerous approaches have been proposed, most of which attempt only "one-shot" reconstruction of the whole network with no intervention from the user, or offer only simple correlation analysis to infer gene dependencies.

Results

We have developed MINER (Microarray Interactive Network Exploration and Representation), an application that combines multivariate non-linear tree learning of individual gene regulatory dependencies, visualisation of these dependencies as both trees and networks, and representation of known biological relationships based on common Gene Ontology annotations. MINER allows biologists to explore the dependencies influencing the expression of individual genes in a gene expression data set in the form of decision, model or regression trees, using their domain knowledge to guide the exploration and formulate hypotheses. Multiple trees can then be summarised in the form of a gene network diagram. MINER is being adopted by several of our collaborators and has already led to the discovery of a new significant regulatory relationship with subsequent experimental validation.

Conclusion

Unlike most gene regulatory network inference methods, MINER allows the user to start from genes of interest and build the network gene-by-gene, incorporating domain expertise in the process. This approach has been used successfully with RNA microarray data but is applicable to other quantitative data produced by high-throughput technologies such as proteomics and "next generation" DNA sequencing.

VisANT 3.5: multi-scale network visualization, analysis and inference based on the gene ontology

Despite its wide usage in biological databases and applications, the role of the gene ontology (GO) in network analysis is usually limited to functional annotation of genes or gene sets with auxiliary information on correlations ignored. Here, we report on new capabilities of VisANT--an integrative software platform for the visualization, mining, analysis and modeling of the biological networks--which extend the application of GO in network visualization, analysis and inference. The new VisANT functions can be classified into three categories. (i) Visualization: a new tree-based browser allows visualization of GO hierarchies. GO terms can be easily dropped into the network to group genes annotated under the term, thereby integrating the hierarchical ontology with the network. This facilitates multi-scale visualization and analysis. (ii) Flexible annotation schema: in addition to conventional methods for annotating network nodes with the most specific functional descriptions available, VisANT also provides functions to annotate genes at any customized level of abstraction. (iii) Finding over-represented GO terms and expression-enriched GO modules: two new algorithms have been implemented as VisANT plugins. One detects over-represented GO annotations in any given sub-network and the other finds the GO categories that are enriched in a specified phenotype or perturbed dataset. Both algorithms take account of network topology (i.e. correlations between genes based on various sources of evidence). VisANT is freely available at http://visant.bu.edu.