NATURAL MICROBIAL COMMUNITY COMPOSITIONS COMPARED BY A BACK-PROPAGATING NEURAL-NETWORK AND CLUSTER-ANALYSIS OF 5S RIBOSOMAL-RNA

The community compositions of free-living and particle-associated bact eria in the Chesapeake Bay estuary mere analyzed by comparing banding patterns of stable low-molecular-weight RNA (SLMW RNA) which include 5 S rRNA and tRNA molecules, By analyzing images of autoradiographs of S LMW RNAs on polyacrylamide gels, band intensities of 5S rRNA were conv erted to binary format for transmission to a back-propagating neural n etwork (NN), The NN was trained to relate binary input to sample stati ons, collection times, positions in the water column, and sample types (e.g., particle-associated versus free-living communities). Dendrogra ms produced by using Euclidean distance and average and Ward's linkage methods on data of three independently trained NNs yielded the follow ing results, (i) Community compositions of Chesapeake Bay mater sample s varied both seasonally and spatially, (ii) Although there was no dif ference in the compositions of free-living and particle-associated bac teria in the summer, these community tapes differed significantly in t he winter, (iii) In the summer, most bag samples had a common 121-nucl eotide 5S rRNA molecule, Although this band occurred in the top water of midbay samples, it did not occur in particle-associated communities of bottom-water samples, (iv) Regardless of the season, midbay sample s had the greatest variety of 5S rRNA sizes, The utility of NNs for in terpreting complex banding patterns in electrophoresis gels was demons trated.