A computational framework for complex disease stratification from multiple large-scale datasets

Abstract

BackgroundMultilevel data integration is becoming a major area of research in systems biology. Within this area, multi-‘omics datasets on complex diseases are becoming more readily available and there is a need to set standards and good practices for integrated analysis of biological, clinical and environmental data. We present a framework to plan and generate single and multi-‘omics signatures of disease states.MethodsThe framework is divided into four major steps: dataset subsetting, feature filtering, ‘omics-based clustering and biomarker identification.ResultsWe illustrate the usefulness of this framework by identifying potential patient clusters based on integrated multi-‘omics signatures in a publicly available ovarian cystadenocarcinoma dataset. The analysis generated a higher number of stable and clinically relevant clusters than previously reported, and enabled the generation of predictive models of patient outcomes.ConclusionsThis framework will help health researchers plan and perform multi-‘omics big data analyses to generate hypotheses and make sense of their rich, diverse and ever growing datasets, to enable implementation of translational P4 medicine.Electronic supplementary materialThe online version of this article (10.1186/s12918-018-0556-z) contains supplementary material, which is available to authorized users.

Prominent release of lipoxygenase generated mediators in a murine house dust mite-induced asthma model

“…Although the U-BIOPRED study contains the largest cohort of severe asthmatics with available urine samples, an increased number of subjects may have provided further power in the consensus clustering and improved quality of the cluster interpretations. A larger data set also offers other data analysis methods based on machine learning to be applied (De Meulder et al, 2018). In this context, it is worth mentioning that unsupervised data analysis methods were initially evaluated, such as principle component analysis (PCA) (Jackson, 1991) or topological data analysis (TDA) (Lum et al, 2013).…”

“…Instead, deviation for ideal stability can be identified by finding the model with minimal area. The area is calculated above and below a pre-defined line in the CDF plot, drawn orthogonal to consensus index 0.5, and can serve as a better estimate of stability (De Meulder et al, 2018).…”

Precision medicine for the discovery of treatable mechanisms in severe asthma

“…Thus, the identification of disease mechanisms should be possible through the statistical analysis of different levels of ‐omics data such as transcriptomics or proteomics. This can be followed by annotation with up‐to‐date ontologies to generate biomarker signatures derived from data collected from a single omics platform (called fingerprint) or those biomarker signatures derived from data collected within multiple ‐ omics platforms (called handprints) . Although multi‐level data integration is a major area of research in systems biology, techniques available to do this remain limited and only a handful of papers have been published in relation to respiratory diseases (Figure ).…”

Needs for Systems Approaches to Better Treat Individuals With Severe Asthma: Predicting Phenotypes and Responses to Treatments

“…Another perspective from omics sciences is the rising concept of "treatable mechanisms" to replace phenotypes. Indeed, omics sciences have highlighted numerous up-or downregulated pathways that represent potential robust biomarkers bearing enough precision necessary for 4P medicine to better model asthma and provide "care and cure" at the level of a single patient and thus move from dissociated fingerprints to integrative handprints (Figures 1, 4) (99,100). Though, one must bear in mind that omic (i.e., molecular phenotype) approach provide association between phenotype and molecular pathways.…”