Laurence Yang

EMILiO: A fast algorithm for genome-scale strain design

Systems-level design of cell metabolism is becoming increasingly important for renewable production of fuels, chemicals, and drugs. Computational models are improving in the accuracy and scope of predictions, but are also growing in complexity. Consequently, efficient and scalable algorithms are increasingly important for strain design. Previous algorithms helped to consolidate the utility of computational modeling in this field. To meet intensifying demands for high-performance strains, both the number and variety of genetic manipulations involved in strain construction are increasing. Existing algorithms have experienced combinatorial increases in computational complexity when applied toward the design of such complex strains. Here, we present EMILiO, a new algorithm that increases the scope of strain design to include reactions with individually optimized fluxes. Unlike existing approaches that would experience an explosion in complexity to solve this problem, we efficiently generated numerous alternate strain designs producing succinate, l-glutamate and l-serine. This was enabled by successive linear programming, a technique new to the area of computational strain design.

Dynamic strain scanning optimization: an efficient strain design strategy for balanced yield, titer, and productivity. DySScO strategy for strain design

BackgroundIn recent years, constraint-based metabolic models have emerged as an important tool for metabolic engineering; a number of computational algorithms have been developed for identifying metabolic engineering strategies where the production of the desired chemical is coupled with the growth of the organism. A caveat of the existing algorithms is that they do not take the bioprocess into consideration; as a result, while the product yield can be optimized using these algorithms, the product titer and productivity cannot be optimized. In order to address this issue, we developed the Dynamic Strain Scanning Optimization (DySScO) strategy, which integrates the Dynamic Flux Balance Analysis (dFBA) method with existing strain algorithms.ResultsIn order to demonstrate the effective of the DySScO strategy, we applied this strategy to the design of Escherichia coli strains targeted for succinate and 1,4-butanediol production respectively. We evaluated consequences of the tradeoff between growth yield and product yield with respect to titer and productivity, and showed that the DySScO strategy is capable of producing strains that balance the product yield, titer, and productivity. In addition, we evaluated the economic viability of the designed strain, and showed that the economic performance of a strain can be strongly affected by the price difference between the product and the feedstock.ConclusionOur study demonstrated that the DySScO strategy is a useful computational tool for designing microbial strains with balanced yield, titer, and productivity, and has potential applications in evaluating the economic performance of the design strains.

Systems biology definition of the core proteome of metabolism and expression is consistent with high-throughput data

Significance Defining a core functional proteome supporting the living process has importance for both developing fundamental understanding of cell functions and for synthetic biology applications. Comparative genomics has been the primary approach to achieve such a definition. Here, we use genome-scale models to define a core proteome that computationally supports basic cellular function. This core proteome for metabolism and protein expression, defined through systems biology methods, is validated and characterized by using multiple disparate data types. Finding the minimal set of gene functions needed to sustain life is of both fundamental and practical importance. Minimal gene lists have been proposed by using comparative genomics-based core proteome definitions. A definition of a core proteome that is supported by empirical data, is understood at the systems-level, and provides a basis for computing essential cell functions is lacking. Here, we use a systems biology-based genome-scale model of metabolism and expression to define a functional core proteome consisting of 356 gene products, accounting for 44% of the Escherichia coli proteome by mass based on proteomics data. This systems biology core proteome includes 212 genes not found in previous comparative genomics-based core proteome definitions, accounts for 65% of known essential genes in E. coli, and has 78% gene function overlap with minimal genomes (Buchnera aphidicola and Mycoplasma genitalium). Based on transcriptomics data across environmental and genetic backgrounds, the systems biology core proteome is significantly enriched in nondifferentially expressed genes and depleted in differentially expressed genes. Compared with the noncore, core gene expression levels are also similar across genetic backgrounds (two times higher Spearman rank correlation) and exhibit significantly more complex transcriptional and posttranscriptional regulatory features (40% more transcription start sites per gene, 22% longer 5′UTR). Thus, genome-scale systems biology approaches rigorously identify a functional core proteome needed to support growth. This framework, validated by using high-throughput datasets, facilitates a mechanistic understanding of systems-level core proteome function through in silico models; it de facto defines a paleome.

A bilevel optimization algorithm to identify enzymatic capacity constraints in metabolic networks

Constraint-based models of metabolism seldom incorporate tight bounds, or capacity constraints, on intracellular fluxes due to the lack of experimental data. This can sometimes lead to inaccurate growth phenotype predictions. Meanwhile, other forms of data such as fitness profiling data from growth competition experiments have been demonstrated to contain valuable information for elucidating key aspects of the underlying metabolic network. Hence, the optimal capacity constraint identification (OCCI) algorithm is developed to reconcile constraint-based models of metabolism with fitness profiling data by identifying a set of flux capacity constraints that optimally fits a wide array of strains. OCCI is able to identify capacity constraints with considerable accuracy by matching 1155 in silico-generated growth rates using a simplified model of Escherichia coli central carbon metabolism. OCCI is expected to be a useful tool for integrating high-throughput fitness measurements with constraint-based models for elucidating metabolic network capacities.

Quantitative time-course metabolomics in human red blood cells reveal the temperature dependence of human metabolic networks

The temperature dependence of biological processes has been studied at the levels of individual biochemical reactions and organism physiology (e.g. basal metabolic rates) but has not been examined at the metabolic network level. Here, we used a systems biology approach to characterize the temperature dependence of the human red blood cell (RBC) metabolic network between 4 and 37 °C through absolutely quantified exo- and endometabolomics data. We used an Arrhenius-type model (Q10) to describe how the rate of a biochemical process changes with every 10 °C change in temperature. Multivariate statistical analysis of the metabolomics data revealed that the same metabolic network-level trends previously reported for RBCs at 4 °C were conserved but accelerated with increasing temperature. We calculated a median Q10 coefficient of 2.89 ± 1.03, within the expected range of 2–3 for biological processes, for 48 individual metabolite concentrations. We then integrated these metabolomics measurements into a cell-scale metabolic model to study pathway usage, calculating a median Q10 coefficient of 2.73 ± 0.75 for 35 reaction fluxes. The relative fluxes through glycolysis and nucleotide metabolism pathways were consistent across the studied temperature range despite the non-uniform distributions of Q10 coefficients of individual metabolites and reaction fluxes. Together, these results indicate that the rate of change of network-level responses to temperature differences in RBC metabolism is consistent between 4 and 37 °C. More broadly, we provide a baseline characterization of a biochemical network given no transcriptional or translational regulation that can be used to explore the temperature dependence of metabolism.

Characterizing metabolic pathway diversification in the context of perturbation size.

Cell metabolism is an important platform for sustainable biofuel, chemical and pharmaceutical production but its complexity presents a major challenge for scientists and engineers. Although in silico strains have been designed in the past with predicted performances near the theoretical maximum, real-world performance is often sub-optimal. Here, we simulate how strain performance is impacted when subjected to many randomly varying perturbations, including discrepancies between gene expression and in vivo flux, osmotic stress, and substrate uptake perturbations due to concentration gradients in bioreactors. This computational study asks whether robust performance can be achieved by adopting robustness-enhancing mechanisms from naturally evolved organisms-in particular, redundancy. Our study shows that redundancy, typically perceived as a ubiquitous robustness-enhancing strategy in nature, can either improve or undermine robustness depending on the magnitude of the perturbations. We also show that the optimal number of redundant pathways used can be predicted for a given perturbation size.

Biomarkers are used to predict quantitative metabolite concentration profiles in human red blood cells

Deep-coverage metabolomic profiling has revealed a well-defined development of metabolic decay in human red blood cells (RBCs) under cold storage conditions. A set of extracellular biomarkers has been recently identified that reliably defines the qualitative state of the metabolic network throughout this metabolic decay process. Here, we extend the utility of these biomarkers by using them to quantitatively predict the concentrations of other metabolites in the red blood cell. We are able to accurately predict the concentration profile of 84 of the 91 (92%) measured metabolites (p < 0.05) in RBC metabolism using only measurements of these five biomarkers. The median of prediction errors (symmetric mean absolute percent error) across all metabolites was 13%. The ability to predict numerous metabolite concentrations from a simple set of biomarkers offers the potential for the development of a powerful workflow that could be used to evaluate the metabolic state of a biological system using a minimal set of measurements.

Thermosensitivity of growth is determined by chaperone-mediated proteome reallocation

Significance How do bacteria adapt to the diverse thermal niches on earth? Evidence accumulates in the protein sequence and structural determinants of thermosensitivity and mechanisms by which molecular chaperones aid protein folding. However, a comprehensive understanding of how thermoadaptation is achieved at the systems level is still missing. Here we reconstruct an integrated genome-scale protein-folding network for Escherichia coli, termed FoldME, that couples both contributing factors to the metabolic state of a cell. FoldME simulations reproduce the asymmetrical bacterial temperature response and delineate the multiscale strategies cells use to resist unfolding stresses induced by high temperature and destabilizing mutations in a single gene. The results highlight how global proteome allocation regulates thermoadaptation through balance between chaperones for folding and translational machinery for biosynthesis. Maintenance of a properly folded proteome is critical for bacterial survival at notably different growth temperatures. Understanding the molecular basis of thermoadaptation has progressed in two main directions, the sequence and structural basis of protein thermostability and the mechanistic principles of protein quality control assisted by chaperones. Yet we do not fully understand how structural integrity of the entire proteome is maintained under stress and how it affects cellular fitness. To address this challenge, we reconstruct a genome-scale protein-folding network for Escherichia coli and formulate a computational model, FoldME, that provides statistical descriptions of multiscale cellular response consistent with many datasets. FoldME simulations show (i) that the chaperones act as a system when they respond to unfolding stress rather than achieving efficient folding of any single component of the proteome, (ii) how the proteome is globally balanced between chaperones for folding and the complex machinery synthesizing the proteins in response to perturbation, (iii) how this balancing determines growth rate dependence on temperature and is achieved through nonspecific regulation, and (iv) how thermal instability of the individual protein affects the overall functional state of the proteome. Overall, these results expand our view of cellular regulation, from targeted specific control mechanisms to global regulation through a web of nonspecific competing interactions that modulate the optimal reallocation of cellular resources. The methodology developed in this study enables genome-scale integration of environment-dependent protein properties and a proteome-wide study of cellular stress responses.

Principles of proteome allocation are revealed using proteomic data and genome-scale models

Integrating omics data to refine or make context-specific models is an active field of constraint-based modeling. Proteomics now cover over 95% of the Escherichia coli proteome by mass. Genome-scale models of Metabolism and macromolecular Expression (ME) compute proteome allocation linked to metabolism and fitness. Using proteomics data, we formulated allocation constraints for key proteome sectors in the ME model. The resulting calibrated model effectively computed the “generalist” (wild-type) E. coli proteome and phenotype across diverse growth environments. Across 15 growth conditions, prediction errors for growth rate and metabolic fluxes were 69% and 14% lower, respectively. The sector-constrained ME model thus represents a generalist ME model reflecting both growth rate maximization and “hedging” against uncertain environments and stresses, as indicated by significant enrichment of these sectors for the general stress response sigma factor σS. Finally, the sector constraints represent a general formalism for integrating omics data from any experimental condition into constraint-based ME models. The constraints can be fine-grained (individual proteins) or coarse-grained (functionally-related protein groups) as demonstrated here. This flexible formalism provides an accessible approach for narrowing the gap between the complexity captured by omics data and governing principles of proteome allocation described by systems-level models.

Global transcriptional regulatory network for Escherichia coli robustly connects gene expression to transcription factor activities

Significance While the transcriptional regulatory network (TRN) of Escherichia coli has expanded considerably in recent years through new chromatin immunoprecipitation (ChIP) data, an open question remains: Does the global TRN, reconstructed by combining ChIP data for individual transcriptions factors, consistently explain observed differential gene expression? We have reconstructed a high-confidence TRN, determined its consistency with transcriptomics and predictive capabilities across multiple conditions, extracted 10 functional regulatory modules, and characterized this network at the sequence and structural levels. Our multiomics algorithmic pipeline is expected to facilitate rigorous validation and prioritization of experiments to elucidate TRNs in other bacteria. Transcriptional regulatory networks (TRNs) have been studied intensely for >25 y. Yet, even for the Escherichia coli TRN—probably the best characterized TRN—several questions remain. Here, we address three questions: (i) How complete is our knowledge of the E. coli TRN; (ii) how well can we predict gene expression using this TRN; and (iii) how robust is our understanding of the TRN? First, we reconstructed a high-confidence TRN (hiTRN) consisting of 147 transcription factors (TFs) regulating 1,538 transcription units (TUs) encoding 1,764 genes. The 3,797 high-confidence regulatory interactions were collected from published, validated chromatin immunoprecipitation (ChIP) data and RegulonDB. For 21 different TF knockouts, up to 63% of the differentially expressed genes in the hiTRN were traced to the knocked-out TF through regulatory cascades. Second, we trained supervised machine learning algorithms to predict the expression of 1,364 TUs given TF activities using 441 samples. The algorithms accurately predicted condition-specific expression for 86% (1,174 of 1,364) of the TUs, while 193 TUs (14%) were predicted better than random TRNs. Third, we identified 10 regulatory modules whose definitions were robust against changes to the TRN or expression compendium. Using surrogate variable analysis, we also identified three unmodeled factors that systematically influenced gene expression. Our computational workflow comprehensively characterizes the predictive capabilities and systems-level functions of an organism’s TRN from disparate data types.