A multi-omics data simulator for complex disease studies and its application to evaluate multi-omics data analysis methods for disease classification

An integrative multi-omics analysis approach that combines multiple types of omics data including genomics, epigenomics, transcriptomics, proteomics, metabolomics, and microbiomics, has become increasing popular for understanding the pathophysiology of complex diseases. Although many multi-omics analysis methods have been developed for complex disease studies, there is no simulation tool that simulates multiple types of omics data and models their relationships with disease status. Without such a tool, it is difficult to evaluate the multi-omics analysis methods on the same scale and to estimate the sample size or power when planning a new multi-omics disease study. We developed a multi-omics data simulator OmicsSIMLA, which simulates genomics (i.e., SNPs and copy number variations), epigenomics (i.e., whole-genome bisulphite sequencing), transcriptomics (i.e., RNA-seq), and proteomics (i.e., normalized reverse phase protein array) data at the whole-genome level. Furthermore, the relationships between different types of omics data, such as meQTLs (SNPs influencing methylation), eQTLs (SNPs influencing gene expression), and eQTM (methylation influencing gene expression), were modeled. More importantly, the relationships between these multi-omics data and the disease status were modeled as well. We used OmicsSIMLA to simulate a multi-omics dataset for breast cancer under a hypothetical disease model, and used the data to compare the performance among existing multi-omics analysis methods in terms of disease classification accuracy and run time. Our results demonstrated that complex disease mechanisms can be simulated by OmicsSIMLA, and a random forest-based method showed the highest prediction accuracy when the multi-omics data were properly normalized.