Flux balance analysis of biological systems: applications and challenges

Systems level modelling and simulations of biological processes are proving to be invaluable in obtaining a quantitative and dynamic perspective of various aspects of cellular function. In particular, constraint-based analyses of metabolic networks have gained considerable popularity for simulating cellular metabolism, of which flux balance analysis (FBA), is most widely used. Unlike mechanistic simulations that depend on accurate kinetic data, which are scarcely available, FBA is based on the principle of conservation of mass in a network, which utilizes the stoichiometric matrix and a biologically relevant objective function to identify optimal reaction flux distributions. FBA has been used to analyse genome-scale reconstructions of several organisms; it has also been used to analyse the effect of perturbations, such as gene deletions or drug inhibitions in silico. This article reviews the usefulness of FBA as a tool for gaining biological insights, advances in methodology enabling integration of regulatory information and thermodynamic constraints, and finally addresses the challenges that lie ahead. Various use scenarios and biological insights obtained from FBA, and applications in fields such metabolic engineering and drug target identification, are also discussed. Genome-scale constraint-based models have an immense potential for building and testing hypotheses, as well as to guide experimentation.

Using a logical model to predict the growth of yeast

Whelan, K. E. and King, R. D. Using a logical model to predict the growth of yeast. BMC Bioinformatics 2008, 9:97

Constraint-based modelling of mixed microbial populations: Application to polyhydroxyalkanoates production

Dissertação para obtenção do Grau de Doutor em Engenharia Química e Bioquímica

targetTB: A target identification pipeline for Mycobacterium tuberculosis through an interactome, reactome and genome-scale structural analysis

Background: Tuberculosis still remains one of the largest killer infectious diseases, warranting the identification of newer targets and drugs. Identification and validation of appropriate targets for designing drugs are critical steps in drug discovery, which are at present major bottle-necks. A majority of drugs in current clinical use for many diseases have been designed without the knowledge of the targets, perhaps because standard methodologies to identify such targets in a high-throughput fashion do not really exist. With different kinds of 'omics' data that are now available, computational approaches can be powerful means of obtaining short-lists of possible targets for further experimental validation. Results: We report a comprehensive in silico target identification pipeline, targetTB, for Mycobacterium tuberculosis. The pipeline incorporates a network analysis of the protein-protein interactome, a flux balance analysis of the reactome, experimentally derived phenotype essentiality data, sequence analyses and a structural assessment of targetability, using novel algorithms recently developed by us. Using flux balance analysis and network analysis, proteins critical for survival of M. tuberculosis are first identified, followed by comparative genomics with the host, finally incorporating a novel structural analysis of the binding sites to assess the feasibility of a protein as a target. Further analyses include correlation with expression data and non-similarity to gut flora proteins as well as 'anti-targets' in the host, leading to the identification of 451 high-confidence targets. Through phylogenetic profiling against 228 pathogen genomes, shortlisted targets have been further explored to identify broad-spectrum antibiotic targets, while also identifying those specific to tuberculosis. Targets that address mycobacterial persistence and drug resistance mechanisms are also analysed. Conclusion: The pipeline developed provides rational schema for drug target identification that are likely to have high rates of success, which is expected to save enormous amounts of money, resources and time in the drug discovery process. A thorough comparison with previously suggested targets in the literature demonstrates the usefulness of the integrated approach used in our study, highlighting the importance of systems-level analyses in particular. The method has the potential to be used as a general strategy for target identification and validation and hence significantly impact most drug discovery programmes.

Heat balance analysis of single stage Gifford-McMahon cycle cryorefrigerator

A heat balance analysis of single stage Gifford-McMahon cycle cryorefrigerator is presented. Ideal refrigeration, actual refrigeration, net refrigeration and the various losses are tabulated. It is observed that pressure-volume losses account for a major fraction of the total losses.

Strategies for efficient disruption of metabolism in Mycobacterium tuberculosis from network analysis

Tuberculosis continues to be a major health challenge, warranting the need for newer strategies for therapeutic intervention and newer approaches to discover them. Here, we report the identification of efficient metabolism disruption strategies by analysis of a reactome network. Protein-protein dependencies at a genome scale are derived from the curated metabolic network, from which insights into the nature and extent of inter-protein and inter-pathway dependencies have been obtained. A functional distance matrix and a subsequent nearness index derived from this information, helps in understanding how the influence of a given protein can pervade to the metabolic network. Thus, the nearness index can be viewed as a metabolic disruptability index, which suggests possible strategies for achieving maximal metabolic disruption by inhibition of the least number of proteins. A greedy approach has been used to identify the most influential singleton, and its combination with the other most pervasive proteins to obtain highly influential pairs, triplets and quadruplets. The effect of deletion of these combinations on cellular metabolism has been studied by flux balance analysis. An obvious outcome of this study is a rational identification of drug targets, to efficiently bring down mycobacterial metabolism.

Enumerating constrained elementary flux vectors of metabolic networks

The concept of elementary vector is generalised to the case where the steady-state space of the metabolic network is not a flux cone but is a general polyhedron due to further inhomogeneous constraints on the flows through some of the reactions. On one hand, this allows to selectively enumerate elementary modes which satisfy certain optimality criteria and this can yield a large computational gain compared with full enumeration. On the other hand, in contrast to the single optimum found by executing a linear program, this enables a comprehensive description of the set of alternate optima often encountered in flux balance analysis. The concepts are illustrated on a metabolic network model of human cardiac mitochondria.

A multi-level multi-scale approach to study essential genes in Mycobacterium tuberculosis

Background: The set of indispensable genes that are required by an organism to grow and sustain life are termed as essential genes. There is a strong interest in identification of the set of essential genes, particularly in pathogens, not only for a better understanding of the pathogen biology, but also for identifying drug targets and the minimal gene set for the organism. Essentiality is inherently a systems property and requires consideration of the system as a whole for their identification. The available experimental approaches capture some aspects but each method comes with its own limitations. Moreover, they do not explain the basis for essentiality in most cases. A powerful prediction method to recognize this gene pool including rationalization of the known essential genes in a given organism would be very useful. Here we describe a multi-level multi-scale approach to identify the essential gene pool in a deadly pathogen, Mycobacterium tuberculosis. Results: The multi-level workflow analyses the bacterial cell by studying (a) genome-wide gene expression profiles to identify the set of genes which show consistent and significant levels of expression in multiple samples of the same condition, (b) indispensability for growth by using gene expression integrated flux balance analysis of a genome-scale metabolic model, (c) importance for maintaining the integrity and flow in a protein-protein interaction network and (d) evolutionary conservation in a set of genomes of the same ecological niche. In the gene pool identified, the functional basis for essentiality has been addressed by studying residue level conservation and the sub-structure at the ligand binding pockets, from which essential amino acid residues in that pocket have also been identified. 283 genes were identified as essential genes with high-confidence. An agreement of about 73.5% is observed with that obtained from the experimental transposon mutagenesis technique. A large proportion of the identified genes belong to the class of intermediary metabolism and respiration. Conclusions: The multi-scale, multi-level approach described can be generally applied to other pathogens as well. The essential gene pool identified form a basis for designing experiments to probe their finer functional roles and also serve as a ready shortlist for identifying drug targets.

Prevalent positive epistasis in Escherichia coli and Saccharomyces cerevisiae metabolic networks

Epistasis refers to the interaction between genes. Although high-throughput epistasis data from model organisms are being generated and used to construct genetic networks(1-3), the extent to which genetic epistasis reflects biologically meaningful interactions remains unclear(4-6). We have addressed this question through in silico mapping of positive and negative epistatic interactions amongst biochemical reactions within the metabolic networks of Escherichia coli and Saccharomyces cerevisiae using flux balance analysis. We found that negative epistasis occurs mainly between nonessential reactions with overlapping functions, whereas positive epistasis usually involves essential reactions, is highly abundant and, unexpectedly, often occurs between reactions without overlapping functions. We offer mechanistic explanations of these findings and experimentally validate them for 61 S. cerevisiae gene pairs.

Studio del network metabolico umano: modelli e applicazioni.

Il lavoro di tesi si basa sull'analisi del database pubblico ReconX, il quale costituisce lo stato dell'arte della ricostruzione in silico del metabolismo umano. Il modello che se ne può estrarre è stato impiegato per simulazioni di metabolismo cellulare in condizioni di ossigeno variabile, valutando l'impatto della carenza di ossigeno sulle reazioni e su i pathways metabolici. Le tecniche impiegate appartengono alla systems biology e sono di tipo bioinformatico e riguardano flux balance analysis e flux variability analysis. I risultati trovati vengono quindi confrontati con la letteratura di settore. E' stato inoltre possibile estrarre dei sotto network dal modello principale, analizzando la rete di connessioni esistente fra i metaboliti e fra le reazioni separatamente. Viene estratto e studiato anche il network di interazione fra pathways, su cui è introdotta una misura ad hoc per valutare la forza di connessione fra i vari processi. Su quest'ultimo network viene anche effettuata un'analisi di tipo stocastico, mostrando che tecniche di tipo markoviano possono essere applicate per indagini diverse da quelle più in uso basate sui flussi. Infine viene mostrata una possibile soluzione per visualizzare graficamente i network metabolici così costituiti.

Velocity distribution for a dilute vibrated granular material

The velocity distribution for a vibrated granular material is determined in the dilute limit where the frequency of particle collisions with the vibrating surface is large compared to the frequency of binary collisions. The particle motion is driven by the source of energy due to particle collisions with the vibrating surface, and two dissipation mechanisms-inelastic collisions and air drag-are considered. In the latter case, a general form for the drag force is assumed. First, the distribution function for the vertical velocity for a single particle colliding with a vibrating surface is determined in the limit where the dissipation during a collision due to inelasticity or between successive collisions due to drag is small compared to the energy of a particle. In addition, two types of amplitude functions for the velocity of the surface, symmetric and asymmetric about zero velocity, are considered. In all cases, differential equations for the distribution of velocities at the vibrating surface are obtained using a flux balance condition in velocity space, and these are solved to determine the distribution function. It is found that the distribution function is a Gaussian distribution when the dissipation is due to inelastic collisions and the amplitude function is symmetric, and the mean square velocity scales as [[U-2](s)/(1 - e(2))], where [U-2](s) is the mean square velocity of the vibrating surface and e is the coefficient of restitution. The distribution function is very different from a Gaussian when the dissipation is due to air drag and the amplitude function is symmetric, and the mean square velocity scales as ([U-2](s)g/mu(m))(1/(m+2)) when the acceleration due to the fluid drag is -mu(m)u(y)\u(y)\(m-1), where g is the acceleration due to gravity. For an asymmetric amplitude function, the distribution function at the vibrating surface is found to be sharply peaked around [+/-2[U](s)/(1-e)] when the dissipation is due to inelastic collisions, and around +/-[(m +2)[U](s)g/mu(m)](1/(m+1)) when the dissipation is due to fluid drag, where [U](s) is the mean velocity of the surface. The distribution functions are compared with numerical simulations of a particle colliding with a vibrating surface, and excellent agreement is found with no adjustable parameters. The distribution function for a two-dimensional vibrated granular material that includes the first effect of binary collisions is determined for the system with dissipation due to inelastic collisions and the amplitude function for the velocity of the vibrating surface is symmetric in the limit delta(I)=(2nr)/(1 - e)much less than 1. Here, n is the number of particles per unit width and r is the particle radius. In this Limit, an asymptotic analysis is used about the Limit where there are no binary collisions. It is found that the distribution function has a power-law divergence proportional to \u(x)\((c delta l-1)) in the limit u(x)-->0, where u(x) is the horizontal velocity. The constant c and the moments of the distribution function are evaluated from the conservation equation in velocity space. It is found that the mean square velocity in the horizontal direction scales as O(delta(I)T), and the nontrivial third moments of the velocity distribution scale as O(delta(I)epsilon(I)T(3/2)) where epsilon(I) = (1 - e)(1/2). Here, T = [2[U2](s)/(1 - e)] is the mean square velocity of the particles.

A general design method for electric machines using magnetic circuit model considering the flux saturation problem

This paper proposes an analytical approach that is generalized for the design of various types of electric machines based on a physical magnetic circuit model. Conventional approaches have been used to predict the behavior of electric machines but have limitations in accurate flux saturation analysis and hence machine dimensioning at the initial design stage. In particular, magnetic saturation is generally ignored or compensated by correction factors in simplified models since it is difficult to determine the flux in each stator tooth for machines with any slot-pole combinations. In this paper, the flux produced by stator winding currents can be calculated accurately and rapidly for each stator tooth using the developed model, taking saturation into account. This aids machine dimensioning without the need for a computationally expensive finite element analysis (FEA). A 48-slot machine operated in induction and doubly-fed modes is used to demonstrate the proposed model. FEA is employed for verification.