Distinguishing prognostic and predictive biomarkers: an information theoretic approach

• DOI: 10.1093/bioinformatics/bty357
• Publication Date: 2018-04-30T19:10:40Z
• Authors: Konstantinos Sech...; Konstantinos Papa...; Paul D Metcalfe; David Svensson; James Weatherall; Gavin Brown
• Keywords: Humans; Precision Medicine; Prognosis; Original Papers; Corrigenda; Biomarkers; 3. Good health

The identification of biomarkers to support decision-making is central personalized medicine, in both clinical and research scenarios. challenge can be seen two halves: identifying predictive markers, which guide the development/use tailored therapies; prognostic other aspects care trial planning, i.e. markers considered as covariates for stratification. Mistakenly assuming a biomarker predictive, when it fact largely (and vice-versa) highly undesirable, result financial, ethical personal consequences. We present framework data-driven ranking on their prognostic/predictive strength, using novel information theoretic method. This approach provides natural algebra discuss quantify individual self-consistent mathematical framework.Our contribution procedure, INFO+, naturally distinguishes versus role each handles higher order interactions. In comprehensive empirical evaluation INFO+ outperforms more complex methods, most notably noise factors dominate, are likely falsely identified they just strongly prognostic. Furthermore, we show that our methods 1-3 orders magnitude faster than competitors, making useful discovery 'big data' Finally, apply identify real trials, introduce new graphical representation greater insight into strength biomarker.R implementations suggested available at https://github.com/sechidis.Supplementary data Bioinformatics online.