In order to carry out early detection and diagnosis of cancer and other diseases, we need a biologically meaningful analytical tool that is capable of dealing with the heterogenous data of the diseased specimens. Analysis of microarray or mass spectrometry data of cancer specimens is very challenging because of their high heterogeneity. The proper mining of data will produce the sought after biomarkers and profiling for early detection, diagnosis, prognosis, and assessment of treatment. We have developed an evolutionary compatible solution to analyze high throughput datasets produced by microarray gene-expression and mass spectrometry proteomic and metabolomic data.
Our analytical paradigm for dealing with heterogenous data, which we termed Phylomics, consists of applying two algorithms to the data. The first is a polarity assessment algorithm, and the second is a maximum parsimony algorithm. Thus, producing a multidimensional classification that is highly predictive and can be used for early detection and other purposes as explained above.
Our success in applying the Phylomics approach is well documented in the scientific literature. Additionally, Phylomics holds significant patents related to early disease detection methods.