GAOPP

An operon consists of a set of adjacent genes on the same DNA strand and is one of the key regulatory features of transcriptional regulatory networks. These genes are co-transcribed into a single-strand mRNA sequence with similar biological function. The ability to create operon maps at the genomic level would be helpful in reconstructing regulatory circuits and to gain a detailed understanding of how genes interact with each other to perform specific functions. However, experimental characterization of operons is expensive, and time consuming, and it cannot be done for all the sequenced genomes. Therefore, computational methods for operon prediction are eagerly needed. Several supervised algorithms using experimental training data, available for few species, and unsupervised algorithms are available to locate multi-gene clusters having operon linkage. These methods are implemented either using some internal software, not available publicly, if available they need to be compiled on appropriate platform.

Here we describe a standalone operon predictor with user friendly GUI requiring a simple input files found at NCBI. This is a PERL implementation of non-supervised operon prediction using Genetic algorithm. Genetic algorithm is chosen because it requires no previous training set and hence independent of any prior experimental data. Analysis revealed that a gene pair having an intergenic distance of -4 to 20 has higher chance to have operon linkage. Similarly 63% of genes with operon linkage share same Functionality group and belong to the same metabolic pathway. The functionality group is determined by COG class. The pathway for each gene is stored at KEGG database.

Citation:

Moharana K, Dikhit M, Sahoo B, Sahoo G, Das P. GAOPP: Operon Prediction in Prokaryotes Using Genetic Algorithm. Curr Bioinform. 2015;10:299−305. Link