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

Prediction and optimization of the laccase-mediated synthesis of the antimicrobial compound iodine (I2)

An artificial neural network (ANN) and genetic algorithm (GA) were applied to improve the laccase-mediated oxidation of iodide (I(-)) to elemental iodine (I2). Biosynthesis of iodine (I2) was studied with a 5-level-4-factor central composite design (CCD). The generated ANN network was mathematically evaluated by several statistical indices and revealed better results than a classical quadratic response surface (RS) model. Determination of the relative significance of model input parameters, ranking the process parameters in order of importance (pH>laccase>mediator>iodide), was performed by sensitivity analysis. ANN-GA methodology was used to optimize the input space of the neural network model to find optimal settings for the laccase-mediated synthesis of iodine. ANN-GA optimized parameters resulted in a 9.9% increase in the conversion rate.

Spatial extraction model for soil environmental quality of anomalous areas in a geographic scale

An approach to establish a soil environmental assessment model was proposed to evaluate the soil environmental quality level. The kriging technique and a self-organizing map (SOM) were integrated to investigate the soil environmental quality in a geographic information system (GIS). In this study, SOM was applied to categorize the data set of nine heavy metals in topsoil. A total of 261 topsoil samples were collected to determine the concentrations of Cu, Pb, Zn, Cd, Ni, Cr, Hg, As, and Mn. The samples were clustered into three classes by SOM and visualized by GIS. The results show that different environmental quality categories are significantly different and that the soil environmental quality assessment model is effective.

The use of a radial basis neural network and genetic algorithm for improving the efficiency of laccase-mediated dye decolourization

A radial basis function neural network (RBF) and genetic algorithm (GA) were applied to improve the efficiency of the oxidative decolourization of the recalcitrant dye Reactive Black 5 (RB 5) by a technical laccase (Trametes spp.) and the natural mediator acetosyringone (ACS). The decolourization of RB 5 in aqueous solution was studied with a 3(4) factorial design including different levels of laccase (2, 100, 200 U L(-1)), acetosyringone (5, 50, 100 μM), pH value (3, 4.5, 6) and incubation time (10, 20, 30 min). The generated RBF network was mathematically evaluated by several statistical indices and revealed better results than a classical quadratic response surface (RS) model. The experimental data showed that within 10 min of incubation time a complete decolourization (>90%) was achieved by using the highest amount of laccase (200 U L(-1)) and acetosyringone (100 μM) at pH 6. By applying the RBF-GA methodology, the efficiency of the laccase-mediated decolourization was improved by minimising the required amount of laccase and acetosyringone by 25% and 21.7% respectively. Complete decolourization (>90%) was obtained within 10 min at the GA-optimised process conditions of laccase (150 U L(-1)) and acetosyringone (78.3 μM) at pH 5.67. These results illustrate that the RBF-GA methodology could be a powerful technique during scale-up studies.

Biodegradation in contaminated aquifers: incorporating microbial/molecular methods

In order to evaluate natural attenuation in contaminated aquifers, there has been a recent recognition that a multidisciplinary approach, incorporating microbial and molecular methods, is required. Observed decreases in contaminant mass and identified footprints of biogeochemical reactions are often used as evidence of intrinsic bioremediation, but characterizing the structure and function of the microbial populations at contaminated sites is needed. In this paper, we review the experimental approaches and microbial methods that are available as tools to evaluate the controls on microbially mediated degradation processes in contaminated aquifers. We discuss the emerging technologies used in biogeochemical studies and present a synthesis of recent studies that serve as models of integrating microbiological approaches with more traditional geochemical and hydrogeologic approaches in order to address important biogeochemical questions about contaminant fate.

Use of unsupervised and supervised artificial neural networks for the identification of lactic acid bacteria on the basis of SDS-PAGE patterns of whole cell proteins

Conventional multivariate statistical techniques (hierarchical cluster analysis, linear discriminant analysis) and unsupervised (Kohonen Self Organizing Map) and supervised (Bayesian network) artificial neural networks were compared for as tools for the classification and identification of 352 SDS-PAGE patterns of whole cell proteins of lactic acid bacteria belonging to 22 species of the genera Lactobacillus, Leuconostoc, Enterococcus, Lactococcus and Streptococcus including 47 reference strains. Electrophoretic data were pre-treated using the logistic weighting function described by Piraino et al. [Piraino, P., Ricciardi, A., Lanorte, M. T., Malkhazova, I., Parente, E., 2002. A new procedure for data reduction in electrophoretic fingerprints of whole-cell proteins. Biotechnol. Lett. 24, 1477-1482]. Hierarchical cluster analysis provided a satisfactory classification of the patterns but was unable to discriminate some species (Leuconostoc, Lb. sakei/Lb. curvatus, Lb. acidophilus/Lb. helveticus, Lb. plantarum/Lb. paraplantarum, Lc. lactis/Lc. raffinolactis). A 7x7 Kohonen self-organizing map (KSOM), trained with the patterns of the reference strains, provided a satisfactory classification of the patterns and was able to discriminate more species than hierarchical cluster analysis. The map was used in predictive mode to identify unknown strains and provided results which in 85.5% of cases matched the classification obtained by hierarchical cluster analysis. Two supervised tools, linear discriminant analysis and a 23:5:2 Bayesian network were proven to be highly effective in the discrimination of SDS-PAGE patterns of Lc. lactis from those of other species. We conclude that data reduction by logistic weighting coupled to traditional multivariate statistical analysis or artificial neural networks provide an effective tool for the classification and identification of lactic acid bacteria on the basis of SDS-PAGE patterns of whole cell proteins.

Tests of rRNA hybridization to microarrays suggest that hybridization characteristics of oligonucleotide probes for species discrimination cannot be predicted

Hybridization of rRNAs to microarrays is a promising approach for prokaryotic and eukaryotic species identification. Typically, the amount of bound target is measured by fluorescent intensity and it is assumed that the signal intensity is directly related to the target concentration. Using thirteen different eukaryotic LSU rRNA target sequences and 7693 short perfect match oligonucleotide probes, we have assessed current approaches for predicting signal intensities by comparing Gibbs free energy (ΔG°) calculations to experimental results. Our evaluation revealed a poor statistical relationship between predicted and actual intensities. Although signal intensities for a given target varied up to 70-fold, none of the predictors were able to fully explain this variation. Also, no combination of different free energy terms, as assessed by principal component and neural network analyses, provided a reliable predictor of hybridization efficiency. We also examined the effects of single-base pair mismatch (MM) (all possible types and positions) on signal intensities of duplexes. We found that the MM effects differ from those that were predicted from solution-based hybridizations. These results recommend against the application of probe design software tools that use thermodynamic parameters to assess probe quality for species identification. Our results imply that the thermodynamic properties of oligonucleotide hybridization are by far not yet understood.

Artificial neural network based identification of Campylobacter species by Fourier transform infrared spectroscopy

Two prototypes of artificial neural network (ANN), multilayer perceptron (MLP), and probabilistic neural network (PNN), were used to analyze infrared (IR) spectral data obtained from intact cells belonging to the species Campylobacter coli and Campylobacter jejuni. In order to establish a consistent identification and typing procedure, mid infrared spectra of these species were obtained by means of a Fourier transform infrared (FT-IR) spectroscope. FT-IR patterns belonging to 26 isolates subclassified into 4 genotypes were pre-processed (normalized, smoothed and derivatized) and grouped into training, verification and test sets. The two architectures tested (PNN, MLP) were developed and trained to identify or leave unassigned a number of IR patterns. Two window ranges (w(4), 1200 to 900 cm(-1); and w(5), 900 to 700 cm(-1)) in the mid IR spectrum were presented as input to the ANN models functioning as pattern recognition systems. No matter the ANN used all the training sets were correctly identified at subspecies level. For the test set, the four-layer MLP network was found to be specially suitable to recognize FT-IR data since it correctly identified 99.16% of unknowns using the w(4) range, and was fully successful in detecting atypical patterns from closely related Campylobacter strains and other bacterial species. The PNN network obtained lower percentages in assignation and rejection. Overall, ANNs constitute an excellent mathematical tool in microbial identification, since they are able to recognize with a high degree of confidence typical as well as atypical FT-IR fingerprints from Campylobacter spp.

Evaluation of gel-pad oligonucleotide microarray technology by using artificial neural networks

Past studies have suggested that thermal dissociation analysis of nucleic acids hybridized to DNA microarrays would improve discrimination among duplex types by scanning through a broad range of stringency conditions. To more fully constrain the utility of this approach using a previously described gel-pad microarray format, artificial neural networks (NNs) were trained to recognize noisy or low-quality data, as might derive from nonspecific fluorescence, poor hybridization, or compromised data collection. The NNs were trained to classify dissociation profiles (melts) into groups based on selected characteristics (e.g., initial signal intensity, area under the curve) using a data set of 21,044 profiles derived from 186 probes hybridized to a study set of RNA extracted from 32 microbes common to the human oral cavity. Three melt profile groups were identified: one group consisted mostly of ideal melt profiles; another group consisted mostly of poor melt profiles; and, the remainder were difficult to classify. Screening of melting profiles of perfect-match hybrids revealed inconsistencies in the form of melting profiles even for identical probes on the same microarray hybridized to same target rRNA. Approximately 18% of perfect-match duplex types were correctly classified as poor. Experimental variability and deviation from ideal melt behavior were shown to be attributable primarily to a method of local background subtraction that was very sensitive to displacement of the grid frames used for image capture (both determined by the image analysis system) and duplexes with low binding constants. Additional results showed that long RNA fragments limit the discriminating power among duplex types.

Coupling of functional gene diversity and geochemical data from environmental samples

Genomic techniques commonly used for assessing distributions of microorganisms in the environment often produce small sample sizes. We investigated artificial neural networks for analyzing the distributions of nitrite reductase genes (nirS and nirK) and two sets of dissimilatory sulfite reductase genes (dsrAB1 and dsrAB2) in small sample sets. Data reduction (to reduce the number of input parameters), cross-validation (to measure the generalization error), weight decay (to adjust model parameters to reduce generalization error), and importance analysis (to determine which variables had the most influence) were useful in developing and interpreting neural network models that could be used to infer relationships between geochemistry and gene distributions. A robust relationship was observed between geochemistry and the frequencies of genes that were not closely related to known dissimilatory sulfite reductase genes (dsrAB2). Uranium and sulfate appeared to be the most related to distribution of two groups of these unusual dsrAB-related genes. For the other three groups, the distributions appeared to be related to pH, nickel, nonpurgeable organic carbon, and total organic carbon. The models relating the geochemical parameters to the distributions of the nirS, nirK, and dsrAB1 genes did not generalize as well as the models for dsrAB2. The data also illustrate the danger (generating a model that has a high generalization error) of not using a validation approach in evaluating the meaningfulness of the fit of linear or nonlinear models to such small sample sizes.

Optimization of single-base-pair mismatch discrimination in oligonucleotide microarrays

The discrimination between perfect-match and single-base-pair-mismatched nucleic acid duplexes was investigated by using oligonucleotide DNA microarrays and nonequilibrium dissociation rates (melting profiles). DNA and RNA versions of two synthetic targets corresponding to the 16S rRNA sequences of Staphylococcus epidermidis (38 nucleotides) and Nitrosomonas eutropha (39 nucleotides) were hybridized to perfect-match probes (18-mer and 19-mer) and to a set of probes having all possible single-base-pair mismatches. The melting profiles of all probe-target duplexes were determined in parallel by using an imposed temperature step gradient. We derived an optimum wash temperature for each probe and target by using a simple formula to calculate a discrimination index for each temperature of the step gradient. This optimum corresponded to the output of an independent analysis using a customized neural network program. These results together provide an experimental and analytical framework for optimizing mismatch discrimination among all probes on a DNA microarray.

A statistical procedure for the analysis of microbial communities based on phenotypic properties of isolates

A novel statistical procedure for the analysis of microbial communities based on phenotypic properties of randomly collected isolates is presented and discussed. The procedure allows the representation of the microbial communities as a set of ellipses in a bidimensional graph. This representation is obtained by the following steps: (a) measurement of a set of binary phenotypic properties for n isolates belonging to k samples, each representing a different community; (b) repeated sampling by bootstrapping of the m samples, thus obtaining, for each community, i subsamples of j isolates; (c) calculation of the frequency of positive results for each test for each subsample; (d) calculation of the matrix of Euclidean distances between the k x i frequency vectors; (e) use of multidimensional scaling (MDS) to obtain a representation in two dimensions of the distance relationships between the frequency vectors; (f) plotting of the 95% confidence ellipses for the i frequency vectors of each of the k communities. By using both simple, synthetic microbial communities, and samples of lactic acid bacteria isolated from natural microbial communities (sourdoughs, compressed yeast, fermented sausages), it was demonstrated that the position and shape of the ellipses are clearly related to the composition of the community, while the relationship between the size of the ellipses and the phenotypical diversity of the community is less straightforward: while communities with very different diversity (measured with the Functional Evenness index and the mean taxonomic distance) had ellipses that were very different in size, there was no strict proportionality between the size of the ellipse and the diversity of the community. Nevertheless, the representation of microbial communities obtained by bootstrapping and multidimensional scaling appears to be superior to the more usual representation based on tabulation of the frequencies of isolates belonging to different clusters.

Ammonia-oxidizing bacteria: a model for molecular microbial ecology

The eutrophication of many ecosystems in recent decades has led to an increased interest in the ecology of nitrogen transformation. Chemolitho-autotrophic ammonia-oxidizing bacteria are responsible for the rate-limiting step of nitrification in a wide variety of environments, making them important in the global cycling of nitrogen. These organisms are unique in their ability to use the conversion of ammonia to nitrite as their sole energy source. Because of the importance of this functional group of bacteria, understanding of their ecology and physiology has become a subject of intense research over recent years. The monophyletic nature of these bacteria in terrestrial environments has facilitated molecular biological approaches in studying their ecology, and progress in this field has been rapid. The ammonia-oxidizing bacteria of the beta-subclass Proteobacteria have become somewhat of a model system within molecular microbial ecology, and this chapter reviews recent progress in our knowledge of their distribution, diversity, and ecology.

Single-base-pair discrimination of terminal mismatches by using oligonucleotide microarrays and neural network analyses

The effects of single-base-pair near-terminal and terminal mismatches on the dissociation temperature (T(d)) and signal intensity of short DNA duplexes were determined by using oligonucleotide microarrays and neural network (NN) analyses. Two perfect-match probes and 29 probes having a single-base-pair mismatch at positions 1 to 5 from the 5' terminus of the probe were designed to target one of two short sequences representing 16S rRNA. Nonequilibrium dissociation rates (i.e., melting profiles) of all probe-target duplexes were determined simultaneously. Analysis of variance revealed that position of the mismatch, type of mismatch, and formamide concentration significantly affected the T(d) and signal intensity. Increasing the concentration of formamide in the washing buffer decreased the T(d) and signal intensity, and it decreased the variability of the signal. Although T(d)s of probe-target duplexes with mismatches in the first or second position were not significantly different from one another, duplexes with mismatches in the third to fifth positions had significantly lower T(d)s than those with mismatches in the first or second position. The trained NNs predicted the T(d) with high accuracies (R(2) = 0.93). However, the NNs predicted the signal intensity only moderately accurately (R(2) = 0.67), presumably due to increased noise in the signal intensity at low formamide concentrations. Sensitivity analysis revealed that the concentration of formamide explained most (75%) of the variability in T(d)s, followed by position of the mismatch (19%) and type of mismatch (6%). The results suggest that position of the mismatch at or near the 5' terminus plays a greater role in determining the T(d) and signal intensity of duplexes than the type of mismatch.

Comparison of microbial and meiofaunal community analyses for determining impact of heavy metal contamination

The impact of long-term heavy metal contamination on soil communities was assessed by a number of methods. These included plate counts of culturable bacteria, community level physiological profiling (CLPP) by analysis of the utilization of multiple carbon sources in BIOLOG plates, community fatty acid methyl ester (C-FAME) profiling and dehydrogenase enzyme activity measurements. These approaches were complemented with microscopic assessments of the diversity of the nematode community. Samples from two sites with different histories of heavy-metal input were assessed. Major differences in microbial and meiofaunal parameters were observed both between and within the sites. There was a large degree of congruence between each of the microbiological approaches. In particular, one sample appeared to be distinguished by a reduction in culturable bacteria (especially pseudomonads), limited response to carbon sources in CLPP, and major differences in extracted fatty acid profiles. The use of multivariate analysis to examine the relationship between microbial and physicochemical measurements revealed that CLPP and plate counts were useful for determining the gross effect of metals on soil microbial communities, whereas proportions of metal-resistant bacteria and dehydrogenase activity differentiated between the two sites. Copper and zinc concentrations and pH all showed significant correlation with the microbial parameters. Nematode community structure was affected to a greater extent by soil pH than by metal content, but the within-site rankings were the same as those achieved for microbiological analyses. The use of these methods for field evaluation of the impact of industrial pollution may be possible provided care is taken when interpreting the data.

Comparison of statistical methods for identification of Streptococcus thermophilus, Enterococcus faecalis, and Enterococcus faecium from randomly amplified polymorphic DNA patterns

Thermophilic streptococci play an important role in the manufacture of many European cheeses, and a rapid and reliable method for their identification is needed. Randomly amplified polymorphic DNA (RAPD) PCR (RAPD-PCR) with two different primers coupled to hierarchical cluster analysis has proven to be a powerful tool for the classification and typing of Streptococcus thermophilus, Enterococcus faecium, and Enterococcus faecalis (G. Moschetti, G. Blaiotta, M. Aponte, P. Catzeddu, F. Villani, P. Deiana, and S. Coppola, J. Appl. Microbiol. 85:25-36, 1998). In order to develop a fast and inexpensive method for the identification of thermophilic streptococci, RAPD-PCR patterns were generated with a single primer (XD9), and the results were analyzed using artificial neural networks (Multilayer Perceptron, Radial Basis Function network, and Bayesian network) and multivariate statistical techniques (cluster analysis, linear discriminant analysis, and classification trees). Cluster analysis allowed the identification of S. thermophilus but not of enterococci. A Bayesian network proved to be more effective than a Multilayer Perceptron or a Radial Basis Function network for the identification of S. thermophilus, E. faecium, and E. faecalis using simplified RAPD-PCR patterns (obtained by summing the bands in selected areas of the patterns). The Bayesian network also significantly outperformed two multivariate statistical techniques (linear discriminant analysis and classification trees) and proved to be less sensitive to the size of the training set and more robust in the response to patterns belonging to unknown species.

Phenotypic biomonitoring using multivariate flow cytometric analysis of multi-stained microorganisms

A new method for monitoring phenotypic profiles of pure cultures and complex microbial communities was evaluated. The approach was to stain microorganisms with a battery of fluorescent dyes prior to flow cytometry analysis (FCM) and to analyse the data using multivariate methods, including principal component analysis and partial least squares. The FCM method was quantitatively evaluated using different mixtures of pure cultures as well as microbial communities. The results showed that the method could quantitatively and reproducibly resolve both populations and communities of microorganisms with 5% abundance in a diverse microbial background. The feasibility of monitoring complex microbial communities over time during the biodegradation of naphthalene using the FCM method was demonstrated. The biodegradation of naphthalene occurred to differing extents in microcosms representing three different types of aromatic-contaminated groundwater and a sample of bio-basin water. The FCM method distinguished each of these four microbial communities. The phenotypic profiles were compared with genotypic profiles generated by random-amplified polymorphic DNA analysis. The genotypic profiles of the microbial communities described only the microbial composition, and not their functional change, whereas the phenotypic profiles seemed to contain information on both the composition and the functional change of the microorganisms. Furthermore, event analysis of the FCM data showed that microbial communities with initially differing compositions could converge towards a similar composition if they had a capacity for high levels of degradation, whereas microbial communities with similar initial compositions could diverge if they differed in biodegrading ability.