Application of neural computing methods for interpreting phospholipid fatty acid profiles of natural microbial communities

The microbial community compositions of surface and subsurface marine sedim ents and sediments lining burrows of marine polychaetes and hemichordates f rom the North Inlet estuary (near Georgetown, S.C.) were analyzed by compar ing ester-linked phospholipid fatty acid (PLFA) profiles with a back-propag ating neural network (NN). The NNs were trained to relate PLFA inputs to se diment type outputs (e.g., surface, subsurface, and burrow lining) and worm species (e.g., Notomastus lobatus, Balanoglossus aurantiacus, and Branchyo asychus americana). Sensitivity analysis was used to determine which of the 60 PLFAs significantly contributed to training the NN. The NN architecture was optimized by changing the number of hidden neurons and calculating the cross-validation error between predicted and actual outputs of training an d test data. The optimal NN architecture was found to be four hidden neuron s with 60-input neurons representing the 60 PLFAs, and four output neurons coding for both sediment types and worm species. Comparison of cross-valida tion results using NNs and linear discriminant analysis (LDA) revealed that NNs had significantly fewer incorrect classifications (2.7%) than LDA (8.4 %). For the NN cross-validation, both sediment type and worm species had 3 incorrect classifications out of 112. For the LDA cross-validation, sedimen t type and worm species had 7 and 12 incorrect classifications out of 112, respectively. Sensitivity analysis of the trained NNs revealed that 17 fatt y acids explained 50% of variability in the data set. These PLFAs were high ly different among sediments and burrow types, indicating significant diffe rences in the microbiota.