PMkbase (version 1.0): an interactive web-based tool for tracking bacterial metabolic traits using phenotype microarrays made interoperable with sequence information and visualizing/processing PM data.

opm is an R package designed to analyse multidimensional OmniLog® phenotype microarray (PM) data. opm provides management, visualization and statistical analysis of PM data, including curve-parameter estimation and discretization, dedicated and customizable plots, metadata management, automated generation of textual and tabular reports, mapping of substrates to databases, batch conversion of files and export to phylogenetic software in the YAML markup language. opm is distributed under the GPL through the Comprehensive R Archive Network (http://cran.r-project.org/package=opm) along with a comprehensive manual and a user-friendly tutorial. Further information may be found at http://www.dsmz.de/research/microorganisms/projects/. johannes.sikorski@dsmz.de.

DuctApe: A suite for the analysis and correlation of genomic and OmniLog™ Phenotype Microarray data

Phenotypic microarray (PM) is a standardized, high-throughput technology for profiling phenotypes of microorganisms, which allows for characterization on around 2,000 different media conditions. The data generated using PM can be incorporated into genome-scale metabolic models to improve their predictive capability. In addition, a comparison of phenotypic profiles of wild-type and gene knockout mutants can give essential information about gene functions of unknown genes. In this chapter, we present a protocol to refine preconstructed metabolic models using the PM data. Both manual refinement and algorithmic approaches for integrating the PM data into metabolic models have been discussed.

There is an increasing need to calibrate microbial community profiles obtained through next generation sequencing (NGS) with relevant taxonomic identities of the microbes, and to further associate these identities with phenotypic attributes. Phenotype MicroArray (PM) techniques provide a semi-high throughput assay for characterization and monitoring the microbial cellular phenotypes. Here, we present detailed descriptions of two different PM protocols used in our recent studies on fungal endophytes of forest trees, and highlight the benefits and limitations of this technique. We found that the PM approach enables effective screening of substrate utilization by endophytes. However, the technical limitations are multifaceted and the interpretation of the PM data challenging. For the best result, we recommend that the growth conditions for the fungi are carefully standardized. In addition, rigorous replication and control strategies should be employed whether using pre-configured, commercial microwell-plates or in-house designed PM plates for targeted substrate analyses. With these precautions, the PM technique is a valuable tool to characterize the metabolic capabilities of individual endophyte isolates, or successional endophyte communities identified by NGS, allowing a functional interpretation of the taxonomic data. Thus, PM approaches can provide valuable complementary information for NGS studies of fungal endophytes in forest trees.

Construction and analysis of a genome-scale metabolic network for Bacillus licheniformis WX-02

In this paper comparative genome and phenotype microarray analyses of Rhodococcus sp. BCP1 and Rhodococcus opacus R7 were performed. Rhodococcus sp. BCP1 was selected for its ability to grow on short-chain n-alkanes and R. opacus R7 was isolated for its ability to grow on naphthalene and on o-xylene. Results of genome comparison, including BCP1, R7, along with other Rhodococcus reference strains, showed that at least 30% of the genome of each strain presented unique sequences and only 50% of the predicted proteome was shared. To associate genomic features with metabolic capabilities of BCP1 and R7 strains, hundreds of different growth conditions were tested through Phenotype Microarray, by using Biolog plates and plates manually prepared with additional xenobiotic compounds. Around one-third of the surveyed carbon sources was utilized by both strains although R7 generally showed higher metabolic activity values compared to BCP1. Moreover, R7 showed broader range of nitrogen and sulphur sources. Phenotype Microarray data were combined with genomic analysis to genetically support the metabolic features of the two strains. The genome analysis allowed to identify some gene clusters involved in the metabolism of the main tested xenobiotic compounds. Results show that R7 contains multiple genes for the degradation of a large set of aromatic and PAHs compounds, while a lower variability in terms of genes predicted to be involved in aromatic degradation was found in BCP1. This genetic feature can be related to the strong genetic pressure exerted by the two different environment from which the two strains were isolated. According to this, in the BCP1 genome the smo gene cluster involved in the short-chain n-alkanes degradation, is included in one of the unique regions and it is not conserved in the Rhodococcus strains compared in this work. Data obtained underline the great potential of these two Rhodococcus spp. strains for biodegradation and environmental decontamination processes.

Biolog Phenotype Microarray (PM) is a technology allowing simultaneous screening of the metabolic behaviour of bacteria under a large number of different conditions. Bacteria may often undergo several cycles of metabolic activity during a Biolog experiment. We introduce a novel algorithm to identify these metabolic cycles in PM experimental data, thus increasing the potential of PM technology in microbiology. Our method is based on a statistical decomposition of the time-series measurements into a set of growth models. We show that the method is robust to measurement noise and captures accurately the biologically relevant signals from the data. Our implementation is made freely available as a part of an R package for PM data analysis and can be found at www.helsinki.fi/bsg/software/Biolog_Decomposition.

Biolog phenotype microarrays (PMs) enable simultaneous, high throughput analysis of cell cultures in different environments. The output is high-density time-course data showing redox curves (approximating growth) for each experimental condition. The software provided with the Omnilog incubator/reader summarizes each time-course as a single datum, so most of the information is not used. However, the time courses can be extremely varied and often contain detailed qualitative (shape of curve) and quantitative (values of parameters) information. We present a novel, Bayesian approach to estimating parameters from Phenotype Microarray data, fitting growth models using Markov Chain Monte Carlo (MCMC) methods to enable high throughput estimation of important information, including length of lag phase, maximal "growth" rate and maximum output. We find that the Baranyi model for microbial growth is useful for fitting Biolog data. Moreover, we introduce a new growth model that allows for diauxic growth with a lag phase, which is particularly useful where Phenotype Microarrays have been applied to cells grown in complex mixtures of substrates, for example in industrial or biotechnological applications, such as worts in brewing. Our approach provides more useful information from Biolog data than existing, competing methods, and allows for valuable comparisons between data series and across different models.

While the proteins of the phosphoenolpyruvate:carbohydrate phosphotransferase system (carbohydrate PTS) have been shown to regulate numerous targets, little such information is available for the nitrogen-metabolic phosphotransferase system (nitrogen-metabolic PTS). To elucidate the physiological role of the nitrogen-metabolic PTS, we carried out phenotype microarray (PM) analysis with Escherichia coli K-12 strain MG1655 deleted for the ptsP gene encoding the first enzyme of the nitrogen-metabolic PTS. Together with the PM data, growth studies revealed that a ptsN (encoding enzyme IIA(Ntr)) mutant became extremely sensitive to leucine-containing peptides (LCPs), while both ptsP (encoding enzyme I(Ntr)) and ptsO (encoding NPr) mutants were more resistant than wild type. The toxicity of LCPs was found to be due to leucine and the dephospho-form of enzyme IIA(Ntr) was found to be necessary to neutralize leucine toxicity. Further studies showed that the dephospho-form of enzyme IIA(Ntr) is required for derepression of the ilvBN operon encoding acetohydroxy acid synthase I catalysing the first step common to the biosynthesis of the branched-chain amino acids.

Much of what we know about the biological impacts of vesicles (MVs) is derived from Arabidopsis thaliana. Our study focused on vesicles from species in the non-model plant group, Pinus (pine) (P. elliottii, P. radiata, and P. patula × Pinus tec (hybrid)). These plants have tougher tissues and strong, acicular-shaped leaves (needles). Herein, we first developed a protocol to guide effective collection of juice fluid from needles and roots in a clean and efficient manner. The effects of these vesicles were characterized in terms of the global nutrient profile of the pine pitch canker fungus, Fusarium circinatum, generated from growing fungal spores on ~400 substrates embedded across BioLog phenotypic microarray (PM) plates (PM1, PM2A: carbon sources; PM3B: nitrogen sources; PM9: osmolytes/pH; PM24C: chemicals). Our findings revealed that MVs, specifically needle-derived MVs (ndMVs) from P. elliottii, disrupt metabolite assimilation in several important pathways, including carbon and nitrogen metabolism. The PM data were also strongly correlated with observed phenotypic effects, including reduced viability and germination of spores in liquid media, as well as impaired filamentous growth on solid media. Importantly, these MV-induced phenotypic effects were reproducible in other filamentous pathogens (e.g., Botrytis cinerea, Chrysoporthe cubensis and F. graminearum) and during a glasshouse trial conducted with F. circinatum-infected P. elliottii seedlings, demonstrating the stable biological effects of ndMVs. Cumulatively, our results suggest that plant-derived vesicles can disrupt metabolism in pathogenic fungi and, therefore, serve as a cost-effective and sustainable source of novel plant protection molecules.

The ability to relate genomic differences in bacterial species to their variability in expressed phenotypes is one of the most challenging tasks in today's biology. Such task is of paramount importance towards the understanding of biotechnologically relevant pathways and possibly for their manipulation. Fundamental prerequisites are the genome-wide reconstruction of metabolic pathways and a comprehensive measurement of cellular phenotypes. Cellular pathways can be reliably reconstructed using the KEGG database, while the OmniLog™ Phenotype Microarray (PM) technology may be used to measure nearly 2,000 growth conditions over time. However, few computational tools that can directly link PM data with the gene(s) of interest followed by the extraction of information on gene-phenotype correlation are available. In this chapter the use of the DuctApe software suite is presented, which allows the joint analysis of bacterial genomic and phenomic data, highlighting those pathways and reactions most probably associated with phenotypic variability. A case study on four Sinorhizobium meliloti strains is presented; more example datasets are available online.

To elucidate the dependence of gene function on the environmental conditions, we focused on quantitative data from a phenotype microarray (PM) of the wild type and ca. 300 single-gene knockout mutants of Escherichia coli K-12 cultured under various medium conditions. We developed an infinite relational model (IRM) applicable to three-valued relational data and applied it to the PM data. The results of gene ontology (GO) analysis showed that mutants with deletion of the genes that encode the enzymes threonine synthase and methionine synthase exhibited reduced cell growth in medium containing amino acids as a nutrient source. By comparing the number and degree of overlap of clusters enriched in certain GO terms obtained by IRM and other biclustering methods, we confirmed the effectiveness of our IRM for omics data analysis.

Novel insight into antimicrobial resistance and sensitivity phenotypes associated to qac and norA genotypes in Staphylococcus aureus

Genotyping studies of Australian Scedosporium isolates have revealed the strong prevalence of a recently described species: Scedosporium aurantiacum. In addition to occurring in the environment, this fungus is also known to colonise the respiratory tracts of cystic fibrosis (CF) patients. A high throughput Phenotype Microarray (PM) analysis using 94 assorted substrates (sugars, amino acids, hexose-acids and carboxylic acids) was carried out for four isolates exhibiting different levels of virulence, determined using a Galleria mellonella infection model. A significant difference was observed in the substrate utilisation patterns of strains displaying differential virulence. For example, certain sugars such as sucrose (saccharose) were utilised only by low virulence strains whereas some sugar derivatives such as D-turanose promoted respiration only in the more virulent strains. Strains with a higher level of virulence also displayed flexibility and metabolic adaptability at two different temperature conditions tested (28 and 37°C). Phenotype microarray data were integrated with the whole-genome sequence data of S. aurantiacum to reconstruct a pathway map for the metabolism of selected substrates to further elucidate differences between the strains.

In this report, a genome-scale reconstruction of Bacillus subtilis metabolism and its iterative development based on the combination of genomic, biochemical, and physiological information and high-throughput phenotyping experiments is presented. The initial reconstruction was converted into an in silico model and expanded in a four-step iterative fashion. First, network gap analysis was used to identify 48 missing reactions that are needed for growth but were not found in the genome annotation. Second, the computed growth rates under aerobic conditions were compared with high-throughput phenotypic screen data, and the initial in silico model could predict the outcomes qualitatively in 140 of 271 cases considered. Detailed analysis of the incorrect predictions resulted in the addition of 75 reactions to the initial reconstruction, and 200 of 271 cases were correctly computed. Third, in silico computations of the growth phenotypes of knock-out strains were found to be consistent with experimental observations in 720 of 766 cases evaluated. Fourth, the integrated analysis of the large-scale substrate utilization and gene essentiality data with the genome-scale metabolic model revealed the requirement of 80 specific enzymes (transport, 53; intracellular reactions, 27) that were not in the genome annotation. Subsequent sequence analysis resulted in the identification of genes that could be putatively assigned to 13 intracellular enzymes. The final reconstruction accounted for 844 open reading frames and consisted of 1020 metabolic reactions and 988 metabolites. Hence, the in silico model can be used to obtain experimentally verifiable hypothesis on the metabolic functions of various genes.