Metabolic gene cluster are operon-like structures of functionally linked metabolic genes. They are common feature in procaryotes and filamentous fungi. In recent years more than a dozen metabolic gene clusters have been characterised in a variety of plant species. PhytoClust is a computational tool to identify and analyze metabolic gene clusters in plants. The program uses the plant genome as an input and runs a Hidden-Markov-Model based search algorithm to detect co-located enzymes based on user-defined detection criteria. The program is based on antiSMASH as the core detection tool and used all the secondary programs also used in antiSMASH.
PhytoClust takes FASTA or EMBL files as an input for the cluster search. You can chose between a selection of plant genomes already stored on the server or upload your own sequence. To detect gene cluster candidates you have two search options. Firstly, you can perform the search based on a list of currently known gene cluster types from different plant species and use default settings or your own parameters for the cluster range (range in which the respective enzymes are to be found) and flanking region (region in proximity of the cluster range that is searched for additional secondary metabolism enzymes). Alternatively, you can create your own cluster rule by selecting up to four enzymes families from plant secondary metabolism and setting the parameters for the cluster range and flanking region. This enables you to search for complete new cluster types or to test if your gene cluster candidate can be found in other species as well.
You can leave your email address and will receive a notification including a download link once your calculations are finished. Your results will be available for download for 7 days after the calculations are finished.
Once submitted the program searches the genome sequence for the given cluster types. After completion the results can be examined in the browse or downloaded for further offline analysis. Moreover, for a selection of species you can run a co-expression analysis on the detected putative gene clusters. If any of the gene cluster candidates show co-expression the results will be displayed with the respective putative cluster.