List2Net: Linking multiple lists of biological data in a network context.

Motivation: A common task in scientific research is the comparison of lists or sets of diverse biological entities such as biomolecules, ontologies, sequences and expression profiles. Such comparisons rely, one way or another, on calculating a measure of similarity either by means of vector correlation metrics, set operations such as union and intersection, or specific measures to capture, for example, sequence homology. Subsequently, depending on the data type, the results are often visualized using heatmaps, Venn, Euler, or Alluvial diagrams. While most of the abovementioned representations offer simplicity and interpretability, their effectiveness holds only for a limited number of lists and specific data types. Conversely, network representations provide a more versatile approach where data lists are viewed as interconnected nodes, with edges representing pairwise commonality, correlation, or any other similarity metric. Networks can represent an arbitrary number of lists of any data type, offering a holistic perspective and most importantly, enabling analytics for characterizing and discovering novel insights in terms of centralities, clusters and motifs that can exist in such networks. While several tools that implement the translation of lists to the various commonly used diagrams, such as Venn and Euler, have been developed, a similar tool that can parse, analyze the commonalities and generate networks from an arbitrary number of lists of the same or heterogenous content does not exist. Results: To address this gap, we introduce List2Net, a web-based tool that can rapidly process and represent lists in a network context, either in a single-layer or multi-layer mode, facilitating network analysis on multi-source/multi-layer data. Specifically, List2Net can seamlessly handle lists encompassing a wide variety of biological data types, such as named entities or ontologies (e.g., lists containing gene symbols), sequences (e.g., protein/peptide sequences), and numeric data types (e.g., omics-based expression or abundance profiles). Once the data is imported, the tool then (i) calculates the commonalities or correlations (edges) between the lists (nodes) of interest, (ii) generates and renders the network for visualization and analysis and (iii) provides a range of exporting options, including vector, raster format visualization but also the calculated edge lists and metrics in tabular format for further analysis in other tools. List2Net is a fast, lightweight, yet informative application that provides network-based holistic insights into the conditions represented by the lists of interest (e.g., disease-to-disease, gene-to-phenotype, drug-to-disease, etc.). As a case study, we demonstrate the utility of this tool applied on publicly available datasets related to Multiple Sclerosis (MS). Using the tool, we showcase the translation of various ontologies characterizing this specific condition on disease-to-disease subnetworks of neurodegenerative, autoimmune and infectious diseases generated from various levels of information such as genetic variation, genes, proteins, metabolites and phenotypic terms.

Multi-omics data integration and network-based analysis drives a multiplex drug repurposing approach to a shortlist of candidate drugs against COVID-19.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic is undeniably the most severe global health emergency since the 1918 Influenza outbreak. Depending on its evolutionary trajectory, the virus is expected to establish itself as an endemic infectious respiratory disease exhibiting seasonal flare-ups. Therefore, despite the unprecedented rally to reach a vaccine that can offer widespread immunization, it is equally important to reach effective prevention and treatment regimens for coronavirus disease 2019 (COVID-19). Contributing to this effort, we have curated and analyzed multi-source and multi-omics publicly available data from patients, cell lines and databases in order to fuel a multiplex computational drug repurposing approach. We devised a network-based integration of multi-omic data to prioritize the most important genes related to COVID-19 and subsequently re-rank the identified candidate drugs. Our approach resulted in a highly informed integrated drug shortlist by combining structural diversity filtering along with experts' curation and drug-target mapping on the depicted molecular pathways. In addition to the recently proposed drugs that are already generating promising results such as dexamethasone and remdesivir, our list includes inhibitors of Src tyrosine kinase (bosutinib, dasatinib, cytarabine and saracatinib), which appear to be involved in multiple COVID-19 pathophysiological mechanisms. In addition, we highlight specific immunomodulators and anti-inflammatory drugs like dactolisib and methotrexate and inhibitors of histone deacetylase like hydroquinone and vorinostat with potential beneficial effects in their mechanisms of action. Overall, this multiplex drug repurposing approach, developed and utilized herein specifically for SARS-CoV-2, can offer a rapid mapping and drug prioritization against any pathogen-related disease.

Integrating multi-source information on a single network to detect disease-related clusters of molecular mechanisms.

The abundance of available information for each disease from multiple sources (e.g. as genetic, regulatory, metabolic, and protein-protein interaction) constitutes both an advantage and a challenge in identifying disease-specific underlying mechanisms. Integration of multi-source data is a rising topic and a great challenge in precision medicine and is crucial in enhancing disease understanding, identifying meaningful clusters of molecular mechanisms and increasing precision and personalisation towards the goal of Predictive, Preventive and Personalised Medicine (PPPM). The overall aim of this work was to develop a novel network-based integration methodology with the following characteristics: (i) maximise the number of data sources, (ii) utilise holistic approaches to integrate these sources (iii) be simple, flexible and extendable, (iv) be conclusive. Here, we present the case of Alzheimer's disease as a paradigm for illustrating our novel approach. SIGNIFICANCE: In this work we present an integration methodology, which aggregates a large number of the available data sources and types by exploiting the holistic nature of network approaches. It is simple, flexible and extendable generating solid conclusions regarding the molecular mechanisms that underlie the input data. We have illustrated the strength of our proposed methodology using Alzheimer's disease as a paradigm. This method is expected to serve as a stepping-stone for further development of integration methods of multi-source omic-data and to contribute to progress towards the goal of Predictive, Preventive and Personalised Medicine (PPPM). The output of this methodology may act as a reference map of implicated pathways in the disease under investigation, where pathways related to additional omics data from any kind of experiment may be projected. This will increase the precision in the understanding of the disease and may contribute to personalised approaches for patients with different disease-related pathway profile, leading to a more precise, personalised and ideally preventive management of the disease.