Multi-omic data integration enables discovery of hidden biological regularities.

Rapid growth in size and complexity of biological data sets has led to the 'Big Data to Knowledge' challenge. We develop advanced data integration methods for multi-level analysis of genomic, transcriptomic, ribosomal profiling, proteomic and fluxomic data. First, we show that pairwise integration of primary omics data reveals regularities that tie cellular processes together in Escherichia coli: the number of protein molecules made per mRNA transcript and the number of ribosomes required per translated protein molecule. Second, we show that genome-scale models, based on genomic and bibliomic data, enable quantitative synchronization of disparate data types. Integrating omics data with models enabled the discovery of two novel regularities: condition invariant in vivo turnover rates of enzymes and the correlation of protein structural motifs and translational pausing. These regularities can be formally represented in a computable format allowing for coherent interpretation and prediction of fitness and selection that underlies cellular physiology.

BiGG Models: A platform for integrating, standardizing and sharing genome-scale models.

Genome-scale metabolic models are mathematically-structured knowledge bases that can be used to predict metabolic pathway usage and growth phenotypes. Furthermore, they can generate and test hypotheses when integrated with experimental data. To maximize the value of these models, centralized repositories of high-quality models must be established, models must adhere to established standards and model components must be linked to relevant databases. Tools for model visualization further enhance their utility. To meet these needs, we present BiGG Models (http://bigg.ucsd.edu), a completely redesigned Biochemical, Genetic and Genomic knowledge base. BiGG Models contains more than 75 high-quality, manually-curated genome-scale metabolic models. On the website, users can browse, search and visualize models. BiGG Models connects genome-scale models to genome annotations and external databases. Reaction and metabolite identifiers have been standardized across models to conform to community standards and enable rapid comparison across models. Furthermore, BiGG Models provides a comprehensive application programming interface for accessing BiGG Models with modeling and analysis tools. As a resource for highly curated, standardized and accessible models of metabolism, BiGG Models will facilitate diverse systems biology studies and support knowledge-based analysis of diverse experimental data.

Reconstruction and modeling protein translocation and compartmentalization in Escherichia coli at the genome-scale.

BACKGROUND: Membranes play a crucial role in cellular functions. Membranes provide a physical barrier, control the trafficking of substances entering and leaving the cell, and are a major determinant of cellular ultra-structure. In addition, components embedded within the membrane participate in cell signaling, energy transduction, and other critical cellular functions. All these processes must share the limited space in the membrane; thus it represents a notable constraint on cellular functions. Membrane- and location-based processes have not yet been reconstructed and explicitly integrated into genome-scale models. RESULTS: The recent genome-scale model of metabolism and protein expression in Escherichia coli (called a ME-model) computes the complete composition of the proteome required to perform whole cell functions. Here we expand the ME-model to include (1) a reconstruction of protein translocation pathways, (2) assignment of all cellular proteins to one of four compartments (cytoplasm, inner membrane, periplasm, and outer membrane) and a translocation pathway, (3) experimentally determined translocase catalytic and porin diffusion rates, and (4) a novel membrane constraint that reflects cell morphology. Comparison of computations performed with this expanded ME-model, named iJL1678-ME, against available experimental data reveals that the model accurately describes translocation pathway expression and the functional proteome by compartmentalized mass. CONCLUSION: iJL1678-ME enables the computation of cellular phenotypes through an integrated computation of proteome composition, abundance, and activity in four cellular compartments (cytoplasm, periplasm, inner and outer membrane). Reconstruction and validation of the model has demonstrated that the iJL1678-ME is capable of capturing the functional content of membranes, cellular compartment-specific composition, and that it can be utilized to examine the effect of perturbing an expanded set of network components. iJL1678-ME takes a notable step towards the inclusion of cellular ultra-structure in genome-scale models.

Genome-scale metabolic reconstructions of multiple Escherichia coli strains highlight strain-specific adaptations to nutritional environments.

Genome-scale models (GEMs) of metabolism were constructed for 55 fully sequenced Escherichia coli and Shigella strains. The GEMs enable a systems approach to characterizing the pan and core metabolic capabilities of the E. coli species. The majority of pan metabolic content was found to consist of alternate catabolic pathways for unique nutrient sources. The GEMs were then used to systematically analyze growth capabilities in more than 650 different growth-supporting environments. The results show that unique strain-specific metabolic capabilities correspond to pathotypes and environmental niches. Twelve of the GEMs were used to predict growth on six differentiating nutrients, and the predictions were found to agree with 80% of experimental outcomes. Additionally, GEMs were used to predict strain-specific auxotrophies. Twelve of the strains modeled were predicted to be auxotrophic for vitamins niacin (vitamin B3), thiamin (vitamin B1), or folate (vitamin B9). Six of the strains modeled have lost biosynthetic pathways for essential amino acids methionine, tryptophan, or leucine. Genome-scale analysis of multiple strains of a species can thus be used to define the metabolic essence of a microbial species and delineate growth differences that shed light on the adaptation process to a particular microenvironment.

Genome-scale models of metabolism and gene expression extend and refine growth phenotype prediction.

Growth is a fundamental process of life. Growth requirements are well-characterized experimentally for many microbes; however, we lack a unified model for cellular growth. Such a model must be predictive of events at the molecular scale and capable of explaining the high-level behavior of the cell as a whole. Here, we construct an ME-Model for Escherichia coli--a genome-scale model that seamlessly integrates metabolic and gene product expression pathways. The model computes ~80% of the functional proteome (by mass), which is used by the cell to support growth under a given condition. Metabolism and gene expression are interdependent processes that affect and constrain each other. We formalize these constraints and apply the principle of growth optimization to enable the accurate prediction of multi-scale phenotypes, ranging from coarse-grained (growth rate, nutrient uptake, by-product secretion) to fine-grained (metabolic fluxes, gene expression levels). Our results unify many existing principles developed to describe bacterial growth.

COBRApy: COnstraints-Based Reconstruction and Analysis for Python.

BACKGROUND: COnstraint-Based Reconstruction and Analysis (COBRA) methods are widely used for genome-scale modeling of metabolic networks in both prokaryotes and eukaryotes. Due to the successes with metabolism, there is an increasing effort to apply COBRA methods to reconstruct and analyze integrated models of cellular processes. The COBRA Toolbox for MATLAB is a leading software package for genome-scale analysis of metabolism; however, it was not designed to elegantly capture the complexity inherent in integrated biological networks and lacks an integration framework for the multiomics data used in systems biology. The openCOBRA Project is a community effort to promote constraints-based research through the distribution of freely available software. RESULTS: Here, we describe COBRA for Python (COBRApy), a Python package that provides support for basic COBRA methods. COBRApy is designed in an object-oriented fashion that facilitates the representation of the complex biological processes of metabolism and gene expression. COBRApy does not require MATLAB to function; however, it includes an interface to the COBRA Toolbox for MATLAB to facilitate use of legacy codes. For improved performance, COBRApy includes parallel processing support for computationally intensive processes. CONCLUSION: COBRApy is an object-oriented framework designed to meet the computational challenges associated with the next generation of stoichiometric constraint-based models and high-density omics data sets. AVAILABILITY: http://opencobra.sourceforge.net/

The genome organization of Thermotoga maritima reflects its lifestyle.

The generation of genome-scale data is becoming more routine, yet the subsequent analysis of omics data remains a significant challenge. Here, an approach that integrates multiple omics datasets with bioinformatics tools was developed that produces a detailed annotation of several microbial genomic features. This methodology was used to characterize the genome of Thermotoga maritima--a phylogenetically deep-branching, hyperthermophilic bacterium. Experimental data were generated for whole-genome resequencing, transcription start site (TSS) determination, transcriptome profiling, and proteome profiling. These datasets, analyzed in combination with bioinformatics tools, served as a basis for the improvement of gene annotation, the elucidation of transcription units (TUs), the identification of putative non-coding RNAs (ncRNAs), and the determination of promoters and ribosome binding sites. This revealed many distinctive properties of the T. maritima genome organization relative to other bacteria. This genome has a high number of genes per TU (3.3), a paucity of putative ncRNAs (12), and few TUs with multiple TSSs (3.7%). Quantitative analysis of promoters and ribosome binding sites showed increased sequence conservation relative to other bacteria. The 5'UTRs follow an atypical bimodal length distribution comprised of "Short" 5'UTRs (11-17 nt) and "Common" 5'UTRs (26-32 nt). Transcriptional regulation is limited by a lack of intergenic space for the majority of TUs. Lastly, a high fraction of annotated genes are expressed independent of growth state and a linear correlation of mRNA/protein is observed (Pearson r = 0.63, p<2.2 × 10(-16) t-test). These distinctive properties are hypothesized to be a reflection of this organism's hyperthermophilic lifestyle and could yield novel insights into the evolutionary trajectory of microbial life on earth.

Reconciling a Salmonella enterica metabolic model with experimental data confirms that overexpression of the glyoxylate shunt can rescue a lethal ppc deletion mutant.

The in silico reconstruction of metabolic networks has become an effective and useful systems biology approach to predict and explain many different cellular phenotypes. When simulation outputs do not match experimental data, the source of the inconsistency can often be traced to incomplete biological information that is consequently not captured in the model. To address this problem, general approaches continue to be needed that can suggest experimentally testable hypotheses to reconcile inconsistencies between simulation and experimental data. Here, we present such an approach that focuses specifically on correcting cases in which experimental data show a particular gene to be essential but model simulations do not. We use metabolic models to predict efficient compensatory pathways, after which cloning and overexpression of these pathways are performed to investigate whether they restore growth and to help determine why these compensatory pathways are not active in mutant cells. We demonstrate this technique for a ppc knockout of Salmonella enterica serovar Typhimurium; the inability of cells to route flux through the glyoxylate shunt when ppc is removed was correctly identified by our approach as the cause of the discrepancy. These results demonstrate the feasibility of our approach to drive biological discovery while simultaneously refining metabolic network reconstructions.

Network context and selection in the evolution to enzyme specificity.

Enzymes are thought to have evolved highly specific catalytic activities from promiscuous ancestral proteins. By analyzing a genome-scale model of Escherichia coli metabolism, we found that 37% of its enzymes act on a variety of substrates and catalyze 65% of the known metabolic reactions. However, it is not apparent why these generalist enzymes remain. Here, we show that there are marked differences between generalist enzymes and specialist enzymes, known to catalyze a single chemical reaction on one particular substrate in vivo. Specialist enzymes (i) are frequently essential, (ii) maintain higher metabolic flux, and (iii) require more regulation of enzyme activity to control metabolic flux in dynamic environments than do generalist enzymes. Furthermore, these properties are conserved in Archaea and Eukarya. Thus, the metabolic network context and environmental conditions influence enzyme evolution toward high specificity.

Raloxifene attenuates Pseudomonas aeruginosa pyocyanin production and virulence.

There has been growing interest in disrupting bacterial virulence mechanisms as a form of infectious disease control through the use of 'anti-infective' drugs. Pseudomonas aeruginosa is an opportunistic pathogen noted for its intrinsic antibiotic resistance that causes serious infections requiring new therapeutic options. In this study, an analysis of the P. aeruginosa PAO1 deduced proteome was performed to identify pathogen-associated proteins. A computational screening approach was then used to discover drug repurposing opportunities, i.e. identifying approved drugs that bind and potentially disrupt the pathogen-associated protein targets. The selective oestrogen receptor modulator raloxifene, a drug currently used in the prevention of osteoporosis and/or invasive breast cancer in post-menopausal women, was predicted from this screen to bind P. aeruginosa PhzB2. PhzB2 is involved in production of the blue pigment pyocyanin produced via the phenazine biosynthesis pathway. Pyocyanin is toxic to eukaryotic cells and has been shown to play a role in infection in a mouse model, making it an attractive target for anti-infective drug discovery. Raloxifene was found to strongly attenuate P. aeruginosa virulence in a Caenorhabditis elegans model of infection. Treatment of P. aeruginosa wild-type strains PAO1 and PA14 with raloxifene resulted in a dose-dependent reduction in pyocyanin production in vitro; pyocyanin production and virulence were also reduced for a phzB2 insertion mutant. These results suggest that raloxifene may be suitable for further development as a therapeutic for P. aeruginosa infection and that such already approved drugs may be computationally screened and potentially repurposed as novel anti-infective/anti-virulence agents.

In silico method for modelling metabolism and gene product expression at genome scale.

Transcription and translation use raw materials and energy generated metabolically to create the macromolecular machinery responsible for all cellular functions, including metabolism. A biochemically accurate model of molecular biology and metabolism will facilitate comprehensive and quantitative computations of an organism's molecular constitution as a function of genetic and environmental parameters. Here we formulate a model of metabolism and macromolecular expression. Prototyping it using the simple microorganism Thermotoga maritima, we show our model accurately simulates variations in cellular composition and gene expression. Moreover, through in silico comparative transcriptomics, the model allows the discovery of new regulons and improving the genome and transcription unit annotations. Our method presents a framework for investigating molecular biology and cellular physiology in silico and may allow quantitative interpretation of multi-omics data sets in the context of an integrated biochemical description of an organism.

A comprehensive genome-scale reconstruction of Escherichia coli metabolism--2011.

The initial genome-scale reconstruction of the metabolic network of Escherichia coli K-12 MG1655 was assembled in 2000. It has been updated and periodically released since then based on new and curated genomic and biochemical knowledge. An update has now been built, named iJO1366, which accounts for 1366 genes, 2251 metabolic reactions, and 1136 unique metabolites. iJO1366 was (1) updated in part using a new experimental screen of 1075 gene knockout strains, illuminating cases where alternative pathways and isozymes are yet to be discovered, (2) compared with its predecessor and to experimental data sets to confirm that it continues to make accurate phenotypic predictions of growth on different substrates and for gene knockout strains, and (3) mapped to the genomes of all available sequenced E. coli strains, including pathogens, leading to the identification of hundreds of unannotated genes in these organisms. Like its predecessors, the iJO1366 reconstruction is expected to be widely deployed for studying the systems biology of E. coli and for metabolic engineering applications.

An experimentally-supported genome-scale metabolic network reconstruction for Yersinia pestis CO92.

BACKGROUND: Yersinia pestis is a gram-negative bacterium that causes plague, a disease linked historically to the Black Death in Europe during the Middle Ages and to several outbreaks during the modern era. Metabolism in Y. pestis displays remarkable flexibility and robustness, allowing the bacterium to proliferate in both warm-blooded mammalian hosts and cold-blooded insect vectors such as fleas. RESULTS: Here we report a genome-scale reconstruction and mathematical model of metabolism for Y. pestis CO92 and supporting experimental growth and metabolite measurements. The model contains 815 genes, 678 proteins, 963 unique metabolites and 1678 reactions, accurately simulates growth on a range of carbon sources both qualitatively and quantitatively, and identifies gaps in several key biosynthetic pathways and suggests how those gaps might be filled. Furthermore, our model presents hypotheses to explain certain known nutritional requirements characteristic of this strain. CONCLUSIONS: Y. pestis continues to be a dangerous threat to human health during modern times. The Y. pestis genome-scale metabolic reconstruction presented here, which has been benchmarked against experimental data and correctly reproduces known phenotypes, provides an in silico platform with which to investigate the metabolism of this important human pathogen.

Omic data from evolved E. coli are consistent with computed optimal growth from genome-scale models.

After hundreds of generations of adaptive evolution at exponential growth, Escherichia coli grows as predicted using flux balance analysis (FBA) on genome-scale metabolic models (GEMs). However, it is not known whether the predicted pathway usage in FBA solutions is consistent with gene and protein expression in the wild-type and evolved strains. Here, we report that >98% of active reactions from FBA optimal growth solutions are supported by transcriptomic and proteomic data. Moreover, when E. coli adapts to growth rate selective pressure, the evolved strains upregulate genes within the optimal growth predictions, and downregulate genes outside of the optimal growth solutions. In addition, bottlenecks from dosage limitations of computationally predicted essential genes are overcome in the evolved strains. We also identify regulatory processes that may contribute to the development of the optimal growth phenotype in the evolved strains, such as the downregulation of known regulons and stringent response suppression. Thus, differential gene and protein expression from wild-type and adaptively evolved strains supports observed growth phenotype changes, and is consistent with GEM-computed optimal growth states.