close

  
Filter options:

Freebase Commons Common /common

  • Researchers in Prof. Sylvia Plevritis’ laboratory have developed a novel machine learning algorithm for discovering the patterns of biological progression underlying microarray gene expression data. This software, called Sample Progression Discovery (SPD), aims to organize samples in order to identify subsets of genes that are responsible for progressive processes - such as cell cycles, stem cell differentiation, or staging of cancer and other diseases. SPD performs this unsupervised analysis even when the underlying process contains branchpoints or the temporal order is not known. Overall, this is a valuable data analysis tool that can be used to generate biological hypotheses about progressive relationships among samples and to identify candidate regulators of the underlying process. SPD could also be applied to a variety of high-dimensional datasets, including genomic, proteomic and image-based data.Stage of Research The inventors have demonstrated that SPD can correctly identify time order and candidate genes from a variety of microarray data. These progressions include cell cycle time series, B-cell differentiation, mouse embryonic stem cell lineages, and prostate cancer staging.

Comments

Hide