The Application of Bayesian Methods in Cancer Prognosis and Prediction

Abstract

With the development of high-throughput biological techniques, high-dimensional omics data have emerged. These molecular data provide a solid foundation for precision medicine and prognostic prediction of cancer. Bayesian methods contribute to constructing prognostic models with complex relationships in omics and improving performance by introducing different prior distribution, which is suitable for modelling the high-dimensional data involved. Using different omics, several Bayesian hierarchical approaches have been proposed for variable selection and model construction. In particular, the Bayesian methods of multi-omics integration have also been consistently proposed in recent years. Compared with single-omics, multi-omics integration modelling will contribute to improving predictive performance, gaining insights into the underlying mechanisms of tumour occurrence and development, and the discovery of more reliable biomarkers. In this work, we present a review of current proposed Bayesian approaches in prognostic prediction modelling in cancer.

From Omics to Multi-Omics Approaches for In-Depth Analysis of the Molecular Mechanisms of Prostate Cancer

“…Additional portals with deposited omics data include the Genomic Data Commons Data Portal, the International Cancer Genome Consortium, the Cancer Genome Atlas, the Cancer Proteome Atlas, the Cancer Cell Line Encyclopedia, the cBioPortal and the Catalogue of Somatic Mutations in Cancer [ 239 , 249 ]. Clearly, limitations still exist for the use and optimal combination of all the availableinformation due to issues, such as complexity, heterogeneity, lack of harmonization and incompleteness, but advanced data integration strategies are being proposed to improve this [ 241 , 249 ].…”

“…This necessitates the development of appropriate statistical analyses and informatic tools to integrate the accessible data. Algorithms mainly based on multivariate, similarity and network approaches, and on Bayesian consensus clustering, have been proposed [ 239 , 240 , 241 ]. Some methods are restrictive with regard to the types of data that can be used (e.g., some of the network methods that use known interactions between molecules) and others applicable in principle to any combination of datasets.…”

Multi-omics data integration approaches for precision oncology

“…Throughout the last decade several bayesian algorithms have been proposed for model construction. 76 We can mention the integrative Bayesian analysis of genomics data (iBAG). This framework uses hierarchical modeling to combine genes and biomarkers associated with clinical outcomes into one model.…”

Bayesian Network Analysis for Prediction of Unplanned Hospital Readmissions of Cancer Patients with Breakthrough Cancer Pain and Complex Care Needs

“…There is an increased interest in Bayesian statistical inference in public health and medical research. Many applications are in the field of cancer prediction and prognosis [ 17 ], but several pieces of research have been conducted in palliative care [ 18 ] as well as in other clinical settings and scenarios [ 19 , 20 , 21 ]. Interestingly, in large cohort of inpatients ( n = 198,972), Roth et al [ 22 ] investigated HRAs from all causes using Bayesian data-driven analytical methods.…”

“…The assumption is that although models must offer predictive results with excellent performance, often there is the need to provide an accurate estimate of the uncertainty of the prediction [ 20 ]. Thus, these Bayesian approaches are also increasingly used for predictive analysis in machine learning and artificial intelligence [ 17 , 24 ].…”