A Review of Integrative Imputation for Multi-Omics Datasets

Integrative Analysis of Next-Generation Sequencing for Next-Generation Cancer Research toward Artificial Intelligence

The rapid improvement of next-generation sequencing (NGS) technologies and their application in large-scale cohorts in cancer research led to common challenges of big data. It opened a new research area incorporating systems biology and machine learning. As large-scale NGS data accumulated, sophisticated data analysis methods became indispensable. In addition, NGS data have been integrated with systems biology to build better predictive models to determine the characteristics of tumors and tumor subtypes. Therefore, various machine learning algorithms were introduced to identify underlying biological mechanisms. In this work, we review novel technologies developed for NGS data analysis, and we describe how these computational methodologies integrate systems biology and omics data. Subsequently, we discuss how deep neural networks outperform other approaches, the potential of graph neural networks (GNN) in systems biology, and the limitations in NGS biomedical research. To reflect on the various challenges and corresponding computational solutions, we will discuss the following three topics: (i) molecular characteristics, (ii) tumor heterogeneity, and (iii) drug discovery. We conclude that machine learning and network-based approaches can add valuable insights and build highly accurate models. However, a well-informed choice of learning algorithm and biological network information is crucial for the success of each specific research question.

Integration strategies of multi-omics data for machine learning analysis

Increased availability of high-throughput technologies has generated an ever-growing number of omics data that seek to portray many different but complementary biological layers including genomics, epigenomics, transcriptomics, proteomics, and metabolomics. New insight from these data have been obtained by machine learning algorithms that have produced diagnostic and classification biomarkers. Most biomarkers obtained to date however only include one omic measurement at a time and thus do not take full advantage of recent multi-omics experiments that now capture the entire complexity of biological systems.

Multi-omics data integration strategies are needed to combine the complementary knowledge brought by each omics layer. We have summarized the most recent data integration methods/ frameworks into five different integration strategies: early, mixed, intermediate, late and hierarchical. In this mini-review, we focus on challenges and existing multi-omics integration strategies by paying special attention to machine learning applications.

Multi-Omics Approaches in Immunological Research

The immune system plays a vital role in health and disease, and is regulated through a complex interactive network of many different immune cells and mediators. To understand the complexity of the immune system, we propose to apply a multi-omics approach in immunological research. This review provides a complete overview of available methodological approaches for the different omics data layers relevant for immunological research, including genetics, epigenetics, transcriptomics, proteomics, metabolomics, and cellomics. Thereafter, we describe the various methods for data analysis as well as how to integrate different layers of omics data. Finally, we discuss the possible applications of multi-omics studies and opportunities they provide for understanding the complex regulatory networks as well as immune variation in various immune-related diseases.

Coupling Machine Learning and Lipidomics as a Tool to Investigate Metabolic Dysfunction-Associated Fatty Liver Disease. A General Overview

Hepatic biopsy is the gold standard for staging nonalcoholic fatty liver disease (NAFLD). Unfortunately, accessing the liver is invasive, requires a multidisciplinary team and is too expensive to be conducted on large segments of the population. NAFLD starts quietly and can progress until liver damage is irreversible. Given this complex situation, the search for noninvasive alternatives is clinically important. A hallmark of NAFLD progression is the dysregulation in lipid metabolism. In this context, recent advances in the area of machine learning have increased the interest in evaluating whether multi-omics data analysis performed on peripheral blood can enhance human interpretation. In the present review, we show how the use of machine learning can identify sets of lipids as predictive biomarkers of NAFLD progression. This approach could potentially help clinicians to improve the diagnosis accuracy and predict the future risk of the disease. While NAFLD has no effective treatment yet, the key to slowing the progression of the disease may lie in predictive robust biomarkers. Hence, to detect this disease as soon as possible, the use of computational science can help us to make a more accurate and reliable diagnosis. We aimed to provide a general overview for all readers interested in implementing these methods.

A Detailed Catalogue of Multi-Omics Methodologies for Identification of Putative Biomarkers and Causal Molecular Networks in Translational Cancer Research

Recent advances in sequencing and biotechnological methodologies have led to the generation of large volumes of molecular data of different omics layers, such as genomics, transcriptomics, proteomics and metabolomics. Integration of these data with clinical information provides new opportunities to discover how perturbations in biological processes lead to disease. Using data-driven approaches for the integration and interpretation of multi-omics data could stably identify links between structural and functional information and propose causal molecular networks with potential impact on cancer pathophysiology. This knowledge can then be used to improve disease diagnosis, prognosis, prevention, and therapy. This review will summarize and categorize the most current computational methodologies and tools for integration of distinct molecular layers in the context of translational cancer research and personalized therapy. Additionally, the bioinformatics tools Multi-Omics Factor Analysis (MOFA) and netDX will be tested using omics data from public cancer resources, to assess their overall robustness, provide reproducible workflows for gaining biological knowledge from multi-omics data, and to comprehensively understand the significantly perturbed biological entities in distinct cancer types. We show that the performed supervised and unsupervised analyses result in meaningful and novel findings.

How insects protect themselves against combined starvation and pathogen challenges, and the implications for reductionism

An explosion of data has provided detailed information about organisms at the molecular level. For some traits, this information can accurately predict phenotype. However, knowledge of the underlying molecular networks often cannot be used to accurately predict higher order phenomena, such as the response to multiple stressors. This failure raises the question of whether methodological reductionism is sufficient to uncover predictable connections between molecules and phenotype. This question is explored in this paper by examining whether our understanding of the molecular responses to food limitation and pathogens in insects can be used to predict their combined effects. The molecular pathways underlying the response to starvation and pathogen attack in insects demonstrates the complexity of real-world physiological networks. Although known intracellular signaling pathways suggest that food restriction should enhance immune function, a reduction in food availability leads to an increase in some immune components, a decrease in others, and a complex effect on disease resistance in insects such as the caterpillar Manduca sexta. However, our inability to predict the effects of food restriction on disease resistance is likely due to our incomplete knowledge of the intra- and extracellular signaling pathways mediating the response to single or multiple stressors. Moving from molecules to organisms will require novel quantitative, integrative and experimental approaches (e.g. single cell RNAseq). Physiological networks are non-linear, dynamic, highly interconnected and replete with alternative pathways. However, that does not make them impossible to predict, given the appropriate experimental and analytical tools. Such tools are still under development. Therefore, given that molecular data sets are incomplete and analytical tools are still under development, it is premature to conclude that methodological reductionism cannot be used to predict phenotype.