1. Michaels GS, Carr DB, Askenazi M, Fuhrman S, Wen X, Somogyi R: Cluster analysis and data visualization of large-scale gene expression data. Pac Symp Biocomput 1998:42-53.
2. Raychaudhuri S, Stuart JM, Altman RB: Principal components analysis to summarize microarray experiments: application to sporulation time series. Pac Symp Biocomput 2000:455-466.
3. Koza JR, Mydlower JD, Lanza G, Yu J, Keanne MA: Reverse engineering of metabolic pathways from observed data using genetic programming. Pac Symp Biocomput 2001:434-445. Genetic programming allows computer programs to evolve under selective pressure in order to maximize their performance on a given task. This paper is the first to apply these methods to genetic network reconstruction.
4. Linear modeling of genetic networks from experimental data;van Someren;Ismb,2000
5. Hartemink AJ, Gifford DK, Jaakkola TS, Young RA: Using graphical models and genomic expression data to statistically validate models of regulatory networks. Pac Symp Biocomput 2001:422-433. Although large amounts of data are required to build a Bayesian network de novo, it is relatively easy to evaluate the compatibility of a network with a given set of data. The investigators encoded two models for galactose regulation and then scored them against experimental data. They were able to recover the correct network in yeast based on 52 expression arrays that were collected without this question in mind.