Strategy to design explorative probes for functional microarrays used in Metabolic Design. After extraction of potential candidate sequences by BLASTp using query reference protein to compare against concatenated Swiss-Prot and TrEMBL databases, a multiple alignment with selected protein sequences and the reference protein is performed. The next step is in two parts: (i) for each molecular site, amino acids are backtranslated, taking into account all genetic code redundancy to determine a degenerate nucleic consensus sequence, (ii) probes are then extracted from this consensus sequence, according to defined user parameters. The program then searches for all potential cross-hybridizations for each selected probe against the 'Cross-hybridization database' by tBLASTn. Kane's criteria are then checked for all positive results by BLASTn. If Kane's criteria are in agreement with a potential cross-hybridization, the program also checks whether it is a potential member of the targeted enzyme family using a BLASTx comparison against the reference protein. Cross-hybridization results are then clustered by BLASTn, stored and visualized in an output file.