Artificial neural network identification of heterotrophic marine bacteria based on their fatty-acid composition

The traditional approach to biochemical identification of marine fresh isolates requires considerably long culture preparation times and large quantities of expensive materials and reagents, and the results are not very reliable. On the other hand, taxonomy tests based on DNA composition, although sensitive and reliable, require long execution times and high costs. A method is presented for the classification of fatty-acid profiles, extracted from marine bacteria strains, at genus level based on supervised artificial neural networks. The proposed method allows the correct identification of all patterns belonging to the training set and almost all patterns belonging to the test set. Moreover, a quantitative measure of the importance of each fatty acid for bacterial classification is also achieved. This measure allows the determination of a cluster of fatty acids to be controlled with greater care. The results show that the proposed method is reproducible and rapid, so that it can be routinely used in the marine microbiology laboratory to identify fresh isolates.