2008 Jan - June - DETAILS

The integration of nanomaterials (size range 1-100 nm) with biomolecules has opened new vistas in the field of biomedicine. Nanomaterials like polymers, liposomes, metal oxide nanoparticles, micelles etc. have been used to deliver drugs to diseased tissues and for imaging the progression of a disease. My talk will focus on the therapeutic application of nanomaterials to inhibit the action of anthrax toxin and imaging applications to test for the presence of plaques in blood vessels in atherosclerosis.
The first part of my talk will focus on designing polymer-based inhibitors of anthrax toxin. These inhibitors work on the principle of polyvalency, which is defined as the simultaneous interaction of several ligands on one surface with several receptors on the other. These polyvalent inhibitors were constructed by conjugating multiple copies of a peptide to various polymer backbones. A novel synthetic technique was developed to synthesize polymers of well-defined molecular weight to study the effect of polymer size and peptide density on the efficacy of these inhibitors in neutralizing anthrax toxin. These well-defined polyvalent inhibitors were effective in neutralizing anthrax toxin in vivo. In addition, the sequence-activity relationship was analyzed to identify amino acids in the peptide sequence critical for its inhibitory activity. A semi-batch RAFT polymerization technique was developed to synthesize biocompatible activated polymers with narrow molecular weight distribution. Polyvalent inhibitors of anthrax toxin based on these biocompatible scaffolds demonstrated nanomolar potency in neutralizing anthrax toxin in vitro.
The second part of my talk will focus on the design of in vivo imaging agents for atherosclerotic plaques using iron oxide nanoparticles. Atherosclerosis is an inflammatory disease caused by high levels of cholesterol in blood. The accumulation of cholesterol-carrying low density lipoprotein (LDL) in the lining of the blood vessels triggers the formation of plaques in these vessels. The project involves using a high-throughput screening technique called phage display to isolate peptides that bind selectively to plaque tissue. A combination of in vitro and in vivo phage display techniques is being used to isolate these peptides. Results from initial screens will be discussed.

The expression level for a recombinant protein produced by P. pastoris seems to be determined largely by its inherent properties such as amino acid sequence, the tertiary structure and the site for expression. The attempts on increasing the protein expression levels by far are focused on genetic manipulations. Although this is very crucial, there is a scope to improve the productivity of P. Pastoris fermentations by undertaking a systematic program of optimizing the entire fermentation process This work aims at undertaking such a program by focusing on strategy to identify and to characterize trends in the behavior of the system, it can be expected that by addressing the process as a whole, rather than narrowly focusing on the protein expression alone.
P. Pastoris methanol-based fermentations for expression of heterologous proteins are typically done in three phases: (i) batch growth on glycerol to attain high cell density (ii) fed-batch growth on glycerol to derepress AOX system and (iii) fed-batch growth on methanol for protein expression. A common approach in determining the operating protocol is to deal with each phase separately, that is, determine the optimal conditions in a sequential manner. We have considered the entire fermentation as a single process and determine the optimal operating conditions.
The first important step in the optimization of fermentation processes is the quantification of various processes occurring in the fermentor. The detailed modeling of Pichia Pastoris cultivations is useful to obtain a better understanding of the biochemical interactions that determine cell growth and product formation. Moreover, these models can potentially be used to calculate optimal operating regimes for fed-batch cultivations In Pichia Pastoris fermentations; the variables of interest are concentrations of cells, glycerol, methanol, oxygen and recombinant protein. We have developed a simple unstructured semi-stoichiometric model for modeling fed-batch process. As a first step, we have modeled fed-batch process with a minimum number of experiments while at the same time building up process knowledge for further optimization.
As pointed out earlier, there are three important phases in P. Pastoris fermentations and key control variables can be identified as: (i) initial glycerol levels for first batch phase, (ii) glycerol addition rate in the second phase and (iii) methanol and/or glycerol addition rate in the third phase. The objective is to maximize the protein expression at the end of the operation by ensuring enough cell growth. Thus, the optimization problem is a multi-objective multi-variable dynamic optimization problem. We have used a novel search technique based on a Genetic Algorithm (GA), which incorporates certain knowledge generated by the use of optimal control theory. Genetic algorithm has been used for addressing the multi-objective, multi-variable optimization problem of P. pastoris fermentation.

The processing of polymer solutions and melts is ubiquitous in the chemical industry. The development of computational models capable of describing common processing operations, is consequently an activity that is pursued vigorously both in industry and in academia. Unfortunately, nearly all computational models breakdown numerically when attempts to simulate industrially realistic flow rates are made. This problem has come to be known as the ``High Weissenberg Number Problem'' or HWNP. To date, in spite of decades of effort to resolve the problem, its origin is unclear, and progress in achieving practically useful computations has been limited. In this talk, it will be shown in the context of a simple benchmark problem, that restricting attention to a continuum description of polymer flow fails to provide insight into the cause of the HWNP. On the other hand, the recognition that polymers are long chain molecules capable of undergoing a transition from a coiled to a stretched state in the presence of an extensional flow, turns out to be crucial for obtaining a clear understanding of the source of computational difficulties. Using stochastic simulations of an ensemble of polymer molecules subjected to the flow field as a diagnostic tool, it will be shown that currently popular continuum models are fundamentally flawed.

The gating mechanism of the spinach leaf aquaporin (AQP) SoPIP2;1 has been described based on high-resolution structures of the open and closed conformations. Being a fast, highly selective water transporter isolated from plants, SoPIP2;1 can be incorporated into industrial membranes for water-filtration applications. For this, it is important to quantify channel permeability, and to drive the conformational equilibrium of SoPIP2;1 towards a constitutively open state. Molecular dynamics (MD) simulations of the closed and open conformations of the SoPIP2;1 tetramer embedded in POPC lipid bilayers were implemented to calculate single channel permeability constants (pf) . The D-loop (residues 181-200) is longer by four residues in SoPIP2;1 compared to mammalian AQPs. It stabilizes the closed conformation by anchoring to the N-terminus by a network of H-bonds mediated by R190 and D191. The double mutant R190A-D191A and a truncated D-loop mutant (TRUNC) of the closed form were tested for increasing equilibrium water permeability. Two copies of each simulation were run. Both mutants increased the permeability of the closed conformation to levels of the open conformation. A persistent H-bond (occupancy: 0.8) between the side-chains of D191 and S36 was detected in the wild-type closed conformation. The D191-S36 interaction was severed in both mutants, driving the D-loop away from the N-terminus of the protein and opening the water channel. The D191-S36 H-bond was not previously implicated in the gating mechanism. Interestingly, S36 is a conserved residue in the PIP2 family.

The adsorption of proteins and colloids at a liquid-solid interface is a key step in many natural and industrial processes such as filtration, chromatography, protein purification, immunological assays, biosensors, biomineralization and biofouling. The surface of the adsorbent is typically heterogeneous. Many of these applications can benefit from a quantitative knowledge of the amount of solute that is adsorbed as a function of the bulk concentration, ligand density and distribution and solute properties. Although widely applied to experimental data, the Langmuir model may provide a poor description of these adsorption systems if steric exclusion effects are present. We will describe a statistical mechanical approach using a random site model to represent the quenched disorder of the substrate. No complete analytical solutions exist, but satisfactory approximate schemes can be developed.

The ability to accelerate the accumulation of favorable combinations of mutations renders recombination a potent force underlying the emergence of forms of HIV that escape multi-drug therapy and specific host-immune responses. We present a mathematical model that describes the dynamics of the emergence of recombinant forms of HIV following infection with diverse viral genomes. Mimicking recent in vitro experiments, we consider target cells simultaneously exposed to two distinct, homozygous viral populations and construct dynamical equations that predict the time-evolution of populations of uninfected, singly infected, and doubly infected cells, and homozygous, heterozygous, and recombinant viruses. Model predictions capture several recent experimental observations quantitatively and provide insights into the role of recombination in HIV dynamics. From comparisons of data from single round infection experiments with model predictions of the probability with which recombination accumulates distinct mutations present on the two genomic strands in a virion, we estimate that ~8 recombinational strand transfer events occur on average (95% confidence interval: 6-10) during reverse transcription of HIV in T cells. Model predictions of virus and cell dynamics describe the time-evolution and the relative prevalence of various infected cell subpopulations following the onset of infection observed experimentally. Remarkably, model predictions are in quantitative agreement with the experimental scaling relationship that the percentage of cells infected with recombinant genomes is proportional to the percentage of cells coinfected with the two genomes employed at the onset of infection. Our model thus presents an accurate description of the influence of recombination on HIV dynamics in vitro. When distinctions between different viral genomes are ignored, our model reduces to the standard model of viral dynamics, which successfully predicts viral load changes in HIV patients undergoing therapy. Our model may thus serve as a useful framework to predict the emergence of multi-drug resistant forms of HIV in infected individuals.

Approaches for state estimation in chemical processes have to deal with many challenges that include nonlinear dynamics and the constraints and bounds imposed on the process variables. The quality of the reconciled process data plays a vital role in the benefits gained from activities such as performance monitoring, online optimization and control. Techniques like Extended Kalman Filter (EKF) and Nonlinear Dynamic Data Reconciliation (NDDR) have been developed for nonlinear systems to improve the reconciled data quality. There are several problems associated with these techniques: EKF cannot handle constraints and bounds, while the NDDR has high computational cost. To address these issues, a new recursive technique called Recursive Nonlinear Dynamic Data Reconciliation (RNDDR), which combines the merits of both EKF and NDDR was developed. The accuracy of this estimator is limited due to the propagation of the covariance matrix through system linearization. Unscented Kalman filter (UKF) avoids the linearization of the nonlinear equations in the covariance propagation. However, the UKF approach can fail in the presence of bounds and constraints. Unscented Recursive Nonlinear Dynamic Data Reconciliation (URNDDR) combines the merits of both UKF and RNDDR. The incorporation of the UKF ideas in the RNDDR framework results in a large number (due to the need to perform optimization at all the sample points) of small (due to the recursive nature of the formulation) optimization problems that need to be solved. In this talk, a new estimation technique named Recursive Unscented Constrained Estimator (RUCE), which solves only a single optimization problem at each time step will be discussed. It will be shown that the estimates developed from RUCE are better than the RNDDR estimates and possess almost the same accuracy as the URNDDR estimates while requiring a significantly lower computational effort.

Those who have played with sand on the beach or with sugar in their kitchen are aware that a collection of solid grains can behave macroscopically like a liquid and ßow. However, the description of this peculiar fluid still represents a challenge due to the lack of constitutive laws able to describe the rich phenomenology observed. This talk will first review the properties of dry granular flows and will present recent advances in our understanding of their rheological behavior. The success and limits of a simple visco-plastic model recently developed will be presented. The second part of this presentation will discuss the more complex case of mixture of grains and liquid, and will show how the recent progresses obtained in the dry case allow to better understand immersed granular flows such as those encountered in submarine avalanches.

Microbial growth and substrate consumption are driven primarily by the control of gene expression. In the biological literature, this control is attributed entirely to specific molecules that activate or inhibit the genes the role of dynamical(nonlinear) phenomenaistypically ignored. The talk will focus on some of these nonlinear phenomena. I will begin by showing that the molecular mechanisms, by themselves, cannot explain the substrate consumption and growth patterns of cell pop ulations. To high light the role of the nonlinear phenomena, I will then show that a model containing none of the molecular mechanisms captures all the observed growth patterns. These arguments will be illustrated by focusing on the dynamics of the lac operon. The expression of this operon is completely inhibited in the presence of glucose. In the literature, this inhibition is attributed to two molecular mechanisms, namely, catabolite repression and inducer exclusion. I will present data showing that these mechanisms play a very limited role. Moreover, a minimal model accounting only for enzyme induction and dilution, but no catabolite repression and inducer exclusion, predicts the complete repression ofthe lac operon. It also explains many other features of the substrate consumption patterns observed in batch and continuous cultures of bacteria and methylotrophic yeasts.

Excess fluoride in drinking water is a cause for concern in several countries in the world. Community -level solutions are being developed to mitigate the harmful effects of fluoride such as dental and skeletal fluorosis. In the present study, a combined alum and activated alumina(AA) process has been investigated. It is seen that pretreatment with alum extends the time between regeneration of the activated alumina column to the extent of 75% in the volume of treated water was achieved when compared to AA process alone. Though regeneration of the AA column has been well-documented, subtle technical issues have not been reported. This study aims to examine some of the issues involved in regeneration of the AA column such as quantity of acid and alkali required.

Ceramic candles are widely-used for water filtration as they are easily available and inexpensive which make them better suited for house-hold water treatment purposes. In the present study, ceramic candles have been coated with nano-size alumina and nano-size magnesium oxide and tested for their defluoridation capacity. It has been found that nano-size magnesium oxide has higher defluoridation capacity than nano-size alumina.

Batch adsorption has been employed to measure the adsorption capacity of adsorbents. A model to capture the overall picture of the batch adsorption process, obtaining the kinetic parameters involved has been developed. The parameters so obtained can be used for modeling the column(continuous) adsorption process.

In recent years, advances in hardware technology and software methodology have prompted the emergence of powerful new computational approaches as a unique new toolset for understanding properties of complex materials. First-principles techniques rely on quantum-mechanical calculations and are truly predictive, requiring no experimental input. This makes them ideally suited for probing atomistic behavior in a wide variety of chemical and physical environments.

In this talk, I will showcase some of the ways first-principles molecular dynamics can be used to illuminate the interplay between various kinetic processes and thermodynamic factors in solids. By drawing examples from materials of current technological interest, I will explore how computational techniques can reach beyond experiments to unravel the complex pathways and mechanisms that underlie structural phase transitions and rapid diffusion of ionic species. I will also demonstrate how molecular dynamics can be combined with various statistical approaches to offer information about the relevant timescales and kinetics of microscopic dynamical events. Finally, I will discuss some novel ways in which thermodynamic information can be extracted from microscopic particle trajectories, and how such information can be used to reinterpret experimental data and understand macroscopic phenomena.

Fluid dynamical problems formulated in unbounded domains have to be truncated if they have to be solved numerically. This requires artificial boundary conditions (abc) on the truncated domain. We report here (after a brief tutorial on abc) our experience in choosing an abc for simulating a singular temperature distribution of the air very near the ground (Ramdas layer) during calm clear nights. This is validated by comparing the numerical simulation with an approximate analytical solution of a simple singularly perturbed linear partial differential equation that is derived from the full blown model.

Centrifugal Contactors are used in the industry today primarily because of their ease of maintenance, high efficiency, low retention volume and small size. In these devices, a single rotor is placed inside a stationary casing, thereby creating an annular region and rotor region. Shear field in the annular region creates a fine dispersion, which then finds its way to the rotor region where the centrifugal force field separates it into its constituent phases.

My project aims at understanding the fluid dynamics in the Rotor zone of the Centrifugal Contactor via computational fluid dynamics (CFD) simulations. The two-phase flow in the Rotor zone has been simulated using the Volume of Fluid model. In the talk, I will present the CFD simulation results for a Rotating Cylinder (no flow case), effect of angular momentum on the flow systems and the effect of viscosity on the pumping capacity of the rotor.

Cell adhesion to a solid substrate is necessary for its normal function. It is possible to engineer cell behavior by controlling the properties of the substrate. Cells adhere to and spread on substrates by assembling multi-protein complexes called focal adhesions. These complexes are physically linked to tension-generating actin stress fibers inside the cell, allowing transfer of intracellular force to the substrate. How focal adhesions and stress fibers are assembled by cells is not well-understood. I will discuss our recent work on measuring the kinetics of intermolecular interactions in focal adhesions formed by living cells. This work has developed new insight into how tension generated in stress fibers can promote adhesion assembly. I will discuss our recent progress in characterizing stress fiber dynamics and mechanics. Finally, I will conclude with an application of nano-rod substrates to modulate cell adhesion and survival.

The use of pulsed lasers in medicine occurred soon after the invention of the laser. The ability to selectively ablate tissue with limited collateral damage has since led to the widespread use of lasers in surgery. In cell biology, laser microsurgery of intracellular organelles has also been used to understand several cellular processes. In both cases, the mechanism of damage by lasers remains poorly understood due to the ultrafast nature of the damage process. For this purpose, we have developed a time-resolved imaging system capable of capturing events on the nanosecond time-scale with <1 micron resolution. Pulsed laser ablation of ex vivo rat corneas at 532 nm was studied using this system. Shockwave propagation and cavitation bubble dynamics associated with laser ablation could be captured in these thick tissue samples with nanosecond time resolution. Acute biological response of cells to these forces was characterized by confocal fluorescence microscopy to study cell viability and cytoskeletal changes. We observed that cells experiencing large deformations due to cavitation could remain viable with minimal changes to the cytoskeleton. This study provides a detailed picture of physical forces produced during laser ablation and their effects on cells. Cell biology applications using this laser microbeam system have also been developed. Neuronal microsurgery in C. elegans was accomplished using 355 nm pulses with 90% animal viability. The laser axotomy was used to develope an assay that allows characterization of proteins involved in transport. In another application, selective transfection of mouse embryonic cells was performed using laser pulses to porate the cell membrane. This allowed the uptake of exogenous plasmids into the cell. The implications of this technique to generate colonies of differentiated cells will be discussed.

I will discuss two problems in evolutionary biology - the first one concerns the punctuated mode of evolution and the second one addresses the problem of the evolution of sex. I will present some analytical results for these models and discuss their experimental relevance.

In this talk, we will discuss three problems where inertia in the system plays a nontrivial role in determining the hydrodynamic stability at low Reynolds number. First is the case of dewetting of a viscoelastic liquid on a solid substrate. Here, we show that in the absence of inertia, the length and time scales of the instability are indeterminate. Inclusion of inertia regularizes the singularity in the growth rate, and removes the ambiguity in the length scale of the most unstable mode. Secondly, we consider the stability of viscoelastic liquid flow past a soft solid layer. For frequencies large compared to the shear rate of the flow, a new class of unstable modes appear because of the inertia of the fluid, which are absent in the creeping-flow limit. Finally, we consider the stability of Newtonian flow through a (visco)elastic neo-Hookean tube. Here, we show that in the absence of inertia in the system (Reynolds number, Re = 0) the system is stable, while even for Re << 1 (but nonzero), a class of shear waves in the solid are destabilized by the flow. In all the three cases, shear waves in the viscoelastic liquid or the deformable solid provide a new time scale in the system at nonzero Reynolds number, and the limit of small Reynolds number is not physically the same as the limit of zero Reynolds number.

Surfaces of thin viscous and elastic films have been observed to undergo instability in presence of electric field resulting in a pattern formation with characteristic wavelength and aspect ratios. It is interesting to compare this phenomena as a function of intrinsic physical properties of the films. We extend these studies to thin metallic films and demonstrate the utility of these procedures to fabricate polymer-based electronic devices and address questions related to interface and contacts in solar cells and molecular diodes.

All engineering with physical and chemical systems makes intense use of differentiation, which is essentially a local linear approximation. Higher order approximations (Taylor polynomials) are important when the linear one is degenerate. The 20th Century saw great mathematical progress in knowing when linearisation will be deceptive, and how much higher to go. This gave a better grasp on buckling and other sudden changes, both theoretically and in numerical practice and real time control. The talk will illustrate both the mathematics and the way it can structure efficient code.

The surface oscillations in deformable wall are known to induce an instability in the adjacent flow even in the absence of inertia. This instability, if understood completely, can be exploited to generate a well-mixed flow with improved transport coefficients in microfluidic systems. To realise such systems, quantitative knowledge of the critical parameter for the on-set of instability, and the nature of bifurcation in the region of transition point, is essential. With this objective, a major portion of this thesis deals with the stability analysis of flow past a flexible surface.

In the limit of zero Reynolds number, we examine the viscous instability for a neo-Hookean solid augmented to incorporate the viscous dissipation in the solid medium. The influence of solid viscosity on the stability behaviour will be discussed. We show that the bifurcation to finite amplitude states is subcritical when the solid viscosity is ignored. However, for non-zero solid viscosity, the analysis reveals a range of solid viscosity for which the nature of bifurcation is supercritical. At high Reynolds number, we find that the instability is driven to the supercritically stable branch. An asymptotic analysis in the limit of high Reynolds number verifies the numerical results.
The viscous instability for Newtonian fluid is extended to the fluid with finite elasticity. The stabilising influence of flow elasticity and polymer addition will be demonstrated. The role of polymer concentration in transition and the nature of instability are examined for a wide range of Weissenberg number. The results are condensed in a map showing the stability boundaries in a parametric space. In the final study, the stability of a plane Couette flow of dilute polymeric fluid is investigated. Three variants of Oldroyd-B model have been analysed: the classical, the diffusive, and the non-homogeneous Oldroyd-B models. For the first two models, the flow is linearly stable, hence the threshold energy necessary for subcritical transition is estimated. The third variant of Oldroyd-B model accounts for non-homogeneous polymer concentration coupled with the stress field. This model exhibits an instability in the linear analysis. The `concentration mode' becomes unstable when the fluid Weissenberg number exceeds a certain transition value.

Pichia pastoris, rapidly growing yeast, has gained a widespread acceptance as a host for production of a number of recombinant proteins. Recent exciting developments in glycosylation engineering have further propelled Pichia pastoris as an expression host for the manufacture of various vaccines and biopharmaceuticals. While significant advances have been made in genetically engineering the cells, engineering of fermentation systems to develop scalable and cost-effective humanlike therapeutic protein manufacturing systems is a challenging task. The critical design considerations to achieve high cell densities, and thereby high protein yields, by supplement of proper nutrient(s) feeding rates.

Pichia pastoris fermentations for production of glycoprotein hormones such as human chorionic gonadotropin (hCG), follicle stimulating hormone (FSH) and luteinizing hormone (LH). In this talk, we will discuss the systematic experimental and model based approach for development of the optimal strategies for achieving highest protein yields in shortest fermentation time. By undertaking a systematic experimental program, we develop two alternate strategies which result in high protein yields. The quantification of the fermentation process allows us to address the problem as a multiple decision ~V multi-objective optimization problem. A novel computational algorithm, which combines the rigor of the optimal control theory and power of evolutionary search algorithm, is presented. The optimal results and its implications for design of high density fermentations are also discussed.

Abstract: In the post-liberalisation period, our markets have been flooded with global brands from multinational corporations, with a parallel increase in advertising, urging people to buy. But there has been no concomitant widening of consumers' awareness, to protect against spurious or sub-standard goods, or drug formulations that have not passed safety tests in the US / EU but are sold in developing countries. We, as educated, urban consumers, have failed to keep abreast of these developments, and are becoming victims of some dubious commercial practices.

Most of the major plant, factory, process, equipment and tool disruptions are avoidable, and yet preventable fault detection and diagnosis strategies are not the norm in most industries. It is not uncommon to see simple and preventable faults disrupt the operation of an entire integrated manufacturing facility. For example, faults such as malfunctioning sensors or actuators, inoperative alarm systems, poor controller tuning or configuration can render the most sophisticated control systems useless. Such disruptions can cost in the excess of $1 million per day and on the average they rob the plant of 7% of its annual capacity.

Over the last decade the fields of multivariate statistics, controller performance monitoring techniques and Bayesian inference methods have merged to develop powerful sensing (e.g. Particle Filter based tools) and condition-based monitoring systems for predictive fault detection and diagnosis. These methods rely on the notion of sensor fusion whereby data from many sensors or units are combined with process information, such as physical connectivity of process units, to give a holistic picture of health of an integrated plant. Such methods combined with embedded digital intelligence are at a stage where such strategies are being implemented for off-line and on-line deployment.

This presentation will outline the field of sensor fusion - the application of signal processing methods, in the temporal as well as spectral domains, on a multitude and NOT singular sensor signals to detect incipient process abnormality before a catastrophic breakdown is likely to occur. This talk will be complemented with industrial case studies to demonstrate the success of these methods