Implementation of artificial neural networks (ANNs) to analysis of inter-taxa communities of benthic microorganisms and macroinvertebrates in a polluted stream

This study was performed to gain an understanding of the structural and functional relationships between inter-taxa communities (macroinvertebrates as consumers, and microbes as decomposers or preys for the invertebrates) in a polluted stream using artificial neural networks techniques. Sediment samples, carrying microorganisms (eubacteria) and macroinvertebrates, were seasonally collected from similar habitats in streams with different levels of pollution. Microbial community taxa and densities were determined using polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) and 16S rDNA sequence analysis techniques. The identity and density of macroinvertebrates were concurrently determined. In general, differences were observed on grouping by self-organizing map (SOM) in polluted, clean and recovering sites based on the microbial densities, while the community patterns were partly dependent on the sampling period. A Spearman rank order correlation analysis revealed correlations of several eubacterial species with those of macroinvertebrates: a negative correlation was observed between Acidovorax sp. (from polluted sites) and Gammaridae (mostly from the clean site), while Herbaspirillum sp. and Janthinobacterium sp. appeared to have positive correlations with some macroinvertebrate species. The population dynamics of the tolerant texa, Tubificidae and Chironomidae, appeared to be related with changes in the densities of Acidovorax sp. This study revealed community relationships between macroinvertebrates and microorganisms, reflecting the connectivity between the two communities via the food chain. A further physio-ecological and symbiological study on the invertebrate–microorganism relationships will be required to understand the degradation and utilization of detritus in aquatic ecosystems as well as to elucidate the roles of the inter-taxa in the recovery of polluted aquatic environments.