Big Data in Systems Biology

As the availability and variety of biological and clinical data continue to grow in the big data era, the synthesis of information from multi-modal data( i.e. data that spans different types and contexts, including imaging data, genetics, proteomics, metabolomics, phenomics, etc) has become a promising mainstream of research. The goal of this research is to gain a better understanding of the interplay of various biological mechanisms, more accurately identify risk factors, and enhance the prediction of disease onset and prognosis. Over the past decade, numerous statistical/machine learning methods have been proposed and expanded to achieve this aim. However, most methods only focus on a single outcome in one study, for example, genome-wide association studies (GWAS) only consider two endpoints, namely genetic effect on one phenotype, but they may ignore information of essential biological pathways that link them. A single outcome may not provide enough information to study complex disease mechanisms in humans or depict heterogeneity profiles of patients' characteristics and exposure effects. Additionally, risk prediction models based on a single outcome in one study often underperform in underrepresented populations, exacerbating known health disparities. In contrast, integrative analysis, which borrows information from multiple outcomes, is key to recovering underlying pathways in system biology and designing precision medicine strategies for treating patients. These outcomes can be extracted from studies with various perspectives, such as multi-omics outcomes in biological studies, regions of interest in imaging studies, or multimodal data that spans different types and contexts (e.g., text, genetics, or lab tests).

The goal of this Research Topic is to give a platform to discussions addressing how to address the major challenges of analysing multiple outcomes and making robust inferences from multi-omics, single-cell omics, multi-modal imaging data, and others. By leveraging multiple outcomes within one study or across multiple studies, integrative analysis can provide a more comprehensive understanding of disease mechanisms and ultimately lead to more effective treatments for patients.

This research topic seeks manuscripts (Original research, methods, reviews and opinion articles) addressing but not limited to:
- Developing novel methods of;
(1) depicting inter-relationships and networking among multi-omics outcomes
(2) predicting disease onset using multi-modality data
(3) integrating genomics/imaging/phenomics data from multiple studies
(4) data fusion in studying single-cell omics and spatial omics
(5) addressing study heterogeneity and/or batch effects in studying system biology.
- Novel development of statistical and machine learning methods, including integrative analysis, probabilistic graphical learning, data fusion, distributed learning, transfer learning, and others, with broad applications to system biology and/or precision medicine.

Please note: We expect the authors to share code, data and algorithms publicly and encourage authors to validate model predictions by comparing them to experimental data, including published experimental data.