Cyanobacteria are prokaryotic, photosynthetic organisms with long evolution, cosmopolitan distribution and great biodiversity. Many cyanobacteria produce secondary metabolites with hepatotoxic, neurotoxic or dermatotoxic effects, called cyanotoxins. Bloom-forming cyanobacteria can reach high concentrations in water basins and cause death of fish, birds and domestic animals, which determines them as ecological and health risk factors. Their simple morphology and high degree of variability make their taxonomic classification problematic. Currently, the classification of Cyanobacteria is subject to a complete revision. There are several systems for classification of these organisms, but the polyphasic approach, which combines morphological, ultrastructural and biochemical features with a leading role of the molecular-genetic markers, is considered to be most applicable for taxonomy of the cyanoprokaryotes. Searching for and using of new molecular markers different from the widely used 16S rRNA gene sequences results in more precise determination of cyanobacterial strains with unclear taxonomic position. The correct taxonomic position and identification of the cyanobacterial strains is very important for all studies related to the biological activity of cyanobacteria, their biotechnological application or in the management and monitoring of water. The aim of this study was to evaluate whether the Light-Repressed Protein (LRP) can be used as a new molecular marker for taxonomic purposes within Cyanobacteria. We have performed phylogenetic analyses of cyanobacterial strains based on LRP gene sequences and amino acid sequences by using minimum evolution (ME), maximum parsimony (MP), maximum-likelihood (ML), and neighbor-joining (NJ) methods. For comparison, we have used the same strains and methods to generate phylogenetic trees based on the 16S rRNA gene sequences. Our results showed that LRP could be used as a suitable molecular marker in resolving different phylogenetic clades within Cyanobacteria at generic level and above. Phylogenetic trees based on the LRP showed that most of the cyanobacterial species/strains belonging to different genera are grouped in separate clades supported by high bootstrap values. The use of LRP as a molecular marker allows to group cyanobacteria according to their habitat (marine or freshwater). For example, LRP-based phylogeny of Prochlorococcus marinus showed clustering of species according to their ecological adaptation to high or low light intensities. The comparison between the LRP trees and the 16S phylogenetic tree showed that the LRP is more accurate molecular marker in resolving phylogenetic relationships within Cyanobacteria at generic and species level than the conserved 16S rRNA gene sequence. [ABSTRACT FROM AUTHOR]