Software Tool for Complex Biomarker Discovery

We propose development of a feature discovery algorithm for identification of multivariate (or composition) biomarker signatures from high-dimensional omics data. The proposed algorithm operates on sets of heterogeneous features and returns a selected subset of significant features and a continuous quantitative metric that is indicative of a particular state of the data (i.e. presence or absence of a toxic hazard). There is no requirement that all features analyzed be of the same type; therefore, the algorithm allows for input of a mixture of data types. The types of data suitable for analysis may be gene expression values, gene copy numbers, presence of particular mutations or SNPs, or methylation patterns. Initially this algorithm was formulated for mining for complex biomarkers in oncology, and is extended here for the purpose of continuous monitoring of environmental and chemical hazards.