Clustering gene expression patterns.

摘要: Recent advances in biotechnology allow researchers to measure expression levels for thousands of genes simultaneously, across different conditions and over time. Analysis data produced by such experiments offers potential insight into gene function regulatory mechanisms. A key step the analysis is detection groups that manifest similar patterns. The corresponding algorithmic problem cluster multicondition In this paper we describe a novel clustering algorithm was developed data. We define an appropriate stochastic error model on input, prove under model, recovers structure with high probability. running time n-gene dataset O[n2[log(n)]c]. also present practical heuristic based same ideas. implemented its performance demonstrated simulated real data, very promising results.