Frontier Computations in the Structure of Atomic Nuclei
By Hai Ah Nam, Calvin W. Johnson |
| A reliable understanding of nuclear reactions is needed for fields as diverse as astrophysics and stewardship of the nation's weapons stockpile. When possible, scientists experimentally measure the reaction rates of interest, but many reactions occur in environments too hot, dense, and radioactive to be recreated safely in the laboratory. Our work, on first principles calculations of nuclear structure and nuclear reactions using state-of-the-art high performance computing, is part of a larger effort to establish a coherent theoretical |
|
|
| formulation that can be applied to nuclear phenomena and address those critical gaps in our knowledge. |
| |
|
|
Investigations into the Parallelization of the General Curvilinear Ocean Model (GCOM) and its Operation Across Cyberinfrastructure Environments
By Mary P. Thomas, Jose Castillo |
| The General Curvilinear Ocean Model (GCOM) uses a 3-D time-dependent curvilinear ocean model to simulate stratified ocean currents over uneven terrains with irregular bottom topographies in a rotating system. Pressure gradients in most models are calculated using three-dimensional stratified ocean models that use bottom-following sigma coordinates that can lead to large errors near steep bathymetry. Numerical errors associated with the numerical discretization of the horizontal pressure gradient terms contribute to this error. The GCOM model differs significantly from more traditional approaches, where the use of Cartesian coordinates forces the model to simulate terrain as a series of equidistant steps. The GCOM model uses an |
|
|
adaptable curvilinear grid. The benefits of this approach include increased resolution in regions of interest, and has resulted in a more efficient algorithm. The GCOM code has been demonstrated to solve fluid flow dynamics for bodies of water that are parts of systems such as the open ocean, a lake, a bay or an estuary. The GCOM code has been developed by Carlos Torres, Jose Castillo, and others. This research project includes the parallelization of the computational sections and the use of web services and cyberinfrastructure to increase overall performance and utilization by the broader ocean modeling community. The project has the following overall goals:
- Identify which parts of the model can be developed into generalized modules and which need to be specialized to the application
- Develop an end-to-end scientific workflow for the application.
- Parallelize and optimize the existing sequential GCOM model using specialized parallel/grid math libraries and numerical methods to run on both single large clusters as well as SOA (if possible).
- Identify new approaches for optimizing the computational grid generation
- Migrate the application onto HPC resources such as those at the SDSU CSRC, the NSF TeraGrid, and the NSF/DOE Open Sciences Grid (OSG).
- Define a services oriented architecture (SOA) on which to accomplish the workflow and hosting/accessing services.
- Define the storage and archival requirements and the associated services
- Define the web services needed for both internal and external workflows
- Host the services to the general environmental modeling community and integrate them into specific projects. |
| |
|
|
A Multiple Linear Regression Study of the Incidence of Dengue Hemorrhagic Fever in Thailand, 1983-2001
By Karen M Campbell
A multiple linear regression study examined an extensive collection of demographic and environmental covariates in relation to the incidence of Dengue Hemorrhagic Fever in Thailand from 1983 to 2001. Spatial and temporal autocorrelation in the data and correlation between covariates required extensive pre-analysis of the data prior to performing the regression. A stepwise regression model was used to assess the pool of covariates in relation to dengue epidemics. The poster presents an overview of the spatial temporal pattern of dengue incidence in Thailand, a summary of the transmission dynamics of the endemic 4 serotype model of dengue, and key observations from the regression study. |
| |
|
|
Total Variation Based Image and Structure Enhancement for Electron Tomography
By Carlos Bazan, Dr. Peter Blomgren, Dr. Terry Frey |
| The objective of this research, from the standpoint of our contribution to public health, is to develop, implement and integrate modern image processing techniques in order to obtain more accurate mitochondrial structural information from data collected using three-dimensional electron tomography (ET). The larger goal is to boost the understanding of the intricate mitochondrial architecture and its relation to functionality. The work is relevant in particular to the structural biology community and to its contribution to |
|
|
public health through the understanding of biological systems. To date, it is firmly established that mitochondrial function plays an important role in the regulation of apoptosis (programmed cell death). There is also evidence that defects in function may be related to many of the most common diseases of aging, such as Alzheimer dementia, Parkinson's disease, type II diabetes mellitus, stroke, atherosclerotic heart disease and cancer. This belief is founded in the observation that mitochondrial function experiences measurable disturbance and observable morphological changes under these circumstances. Electron tomography has allowed significant advances in the understanding of mitochondrial structures. Despite the recent advancement in imaging hardware and specimen fixation techniques, the interpretation and measurement of the structural architecture of mitochondria depend substantially on the availability of good software tools for filtering, segmenting, measuring and classifying the features of interest. It has been argued in the structural biology community that the image processing methodologies in the three-dimensional ET field are not yet sufficiently developed, so as to correctly extract features and understand spatial relationships in mitochondrial structures. The main motivation for the proposed work is the development of mathematically sound and computationally robust algorithms for the reduction of noise and enhancement of structural information in mitochondrial images. Our research efforts will initially focus on the understanding of the sources and mechanisms which introduce noise (image degradation) in ET process. Correct modeling of the degradation, will allow us to design effective, correct, algorithms for the removal of the noise without negatively impacting fine-detail structural information, and will allow for a posteriori reconstruction of more accurate three-dimensional mitochondrion images. Our noise reduction and reconstruction efforts will employ a partial differential equation (PDE) approach based on the minimization of the total variation norm, which has been utilized very successfully in other areas of computer vision.
Carlos Bazan was supported as a trainee by NIH RoadMap Initiative award R90 DK07015. |
| |
|
|
Computational Study of Tocopherol Regeneration Kinetics
By Andrew Cooksy, Ray Lui |
| A computational study is presented of three aqueous reactions of elementary tocopherones formed from antioxidant activity of alpha-tocopherol (vitamin E): (1) proton-transfer to produce 1-benzopyrylium, (2) subsequent hydrolysis to 2H-1-benzopyran-6(8aH)-one, (3) terminating rearrangement to 2,5-cyclohexadiene-1,4-dione. Calculations along each reaction pathway were first |
|
|
| performed at the BP86/TZVP/DGA1 level with a COSMO solvent model. Additional solvent corrections from COSMO-RS theory were then added. The ion-mediated steps are found to have free-energies of activation under 3 kcal/mol, with kinetics likely to be diffusion-limited. The unimolecular rearrangement is the slow step, with a predicted reaction rate constant of 0.056 /min at 298 K, in excellent agreement with the experimental value of 0.046 /min-1. Kinetic and thermodynamic properties of the ion-mediated steps are shown to rely strongly on the use of an accurate continuum solvation model. Distinct entropy and enthalpy contributions are determined from the temperature-dependence of the predicted free energies. |
| |
|
|
High-Order Modeling of Rupture Propagation Under Several Constitutive Laws
By Otilio J Rojas, Steven Day, Jose Castillo, Luis A Dalguer |
| Precise and realistic numerical modeling of seismic wave and rupture propagation has become essential for investigating earthquake physics and Earth's structure. In this work, we present an accurate 2D numerical method for in-plane rupture propagation in elastic media that considers physics-based constitutive friction laws. Accurate description of the complex dynamics along the rupture path is achieved by a combination of the |
|
|
| traction-at-split-node (TSN) representation of faulting boundary conditions, and a high-order discretization of the coupled TSN-elastodynamics equations. Fourth-order mimetic finite differences are used for spatial differentiation. This new method is applied to the simulation of frictional sliding governed by slip-dependent and rate- and state-dependent constitutive models. Results are assessed in terms of metrics based upon RMS differences in rupture time, final slip, and peak slip rate. |
| |
|
|
Interaction Between Major Pathways for Calcium Reuptake in the Cardiac Cell
By Magda Nemeth, Karin Norgard-Sumnicht, Paul Paolini |
| Two major regulatory systems are responsible for the reuptake of free calcium during relaxation in adult rat myocytes, the SR Calcium pump, coded by the gene SERCA (Atp2a2, Atp2a3), and the Na-Ca exchanger, NCX (genes Slc8a1, Slc8a2). The separation of the contributions made by the two systems is being achieved with the use of RNA interference (RNAi) to suppress the target gene's expression in both adult and neonatal rat myocytes. Transient magnitude and time course changes are measured and compared with gene expression results from quantitative PCR, Western |
|
|
blotting, and microarray analyses. We have developed an algorithm for teasing out the two time courses associated with calcium binding sites from the experimentally measured calcium transients. The transients are measured using a photomultiplier-based fluorescence microscopy system with the calcium-sensitive dye Fluo-4. Our laboratory is also developing a computer simulation to predict the outcome of gene silencing on the other major relaxation process. A decline in SERCA expression is compensated for by an increase in the NCX expression. We will also observe the results performed in the reverse experiment, by down-regulating NCX using RNAi with transfection and electroporation of the cells, and measuring a subsequent up-regulation of SERCA. This experiment can help test the way in which the anti-diabetic drug, rosiglitazone, is able to increase cardiac cell contractile strength and speed. This experiment demonstrates that the two decay curves for SERCA and NCX are affected by the drug treatment and work in a compensatory fashion to accomplish calcium reuptake during myocyte relaxation; comparison of experimental and theoretical approaches to these gene expression studies will clarify the interplay between cell processes controlling short-term contractile behavior.
Magda Nemeth was supported as a trainee by NIH RoadMap Initiative award T90 DK07015. |
| |
|
|
Calcium Pathway Gene Expression Changes Produced by Rosiglitazone in the Neonatal Rat Cardiocyte
By Lynelle Garnica, Kirubel Gebresenbet, Frank Gonzales, Paul Paolini |
| Heart failure is one of the leading causes of death in the United States, and is the primary cause of death in diabetes. Our investigation involved profiling the gene expression changes produced in response to treatment with the drug rosiglitazone (Avandia©, GlaxoSmithKline), a widely prescribed diabetes medication. We previously studied the effect of rosiglitazone on calcium regulation in neonatal rat ventricular myocytes by measurement of calcium transient decay rates and SERCA2 gene expression |
|
|
| changes using RT-PCR. That study showed that rosiglitazone conferred cardioprotective effects. Myocyte contractions were stronger and showed faster relaxation. The present study used RNA microarray analysis for genetic profiling of the cardiomyocytes in response to rosiglitazone over a 48 hour period. Primary ventricular cardiomyocytes were harvested from neonatal rats and plated out in culture media. The RNA of the cells was extracted, purified, and hybridized with cDNA probes. Microarray analysis software packages, CodeLink 4.1 (Amersham Biosciences) and Genesis© (A. Sturn, Graz University, Austria) were used to characterize resulting patterns of gene expression. An initial list of 9,911 genes was reduced to a much smaller set of important genes involving the calcium signaling pathways at tested time intervals (0,1, 2, 24, and 48 hours). K-means cluster analysis showed the general expression patterns of these genes, assigned to cluster groups based on their functional role in calcium regulation. Expression changes were analyzed for thirty-two clusters of genes in the calcium pathway. Observed changes in regulation of the cardiac ryanodine receptor gene, Ryr2, one of the SERCA genes, the DHP receptor gene, and the NCX genes and could be responsible for the improved contractile behavior of rosiglitazone-treated cells. We have most recently focused on a few of these genes of interest - SERCA2, ryanodine, and phospholamban - and performed Western blot and RT-PCR analysis to validate microarray-based changes in levels of protein product for these genes. (Supported in part by NIH/NIGMS SDSU training grants 5T34GM08303 (MARC1), by NIH grant 5 R90 DK071512-02 (Training for a New Interdisciplinary Workforce2), by a grant from the California Metabolic Research Foundation, and by additional support from the Computational Science Research Center3, San Diego State University; and The Rees Stealy Research Foundation, 2001 Fourth Avenue, San Diego, CA 92101.) |
| |
|
|
Mitochondrial Stereology from the Statistics of Sections
By Danny Flynn, Arlette Baljon, James Nulton, Peter Salamon |
| During apoptosis mitochondria within the cell fragment from tubular shapes to swollen prolate spheroids. Such mitochondria resemble a spatial distribution of prolate spheroids of various shapes, sizes and orientations. The cell is studied by making electron micrographs which amount to a series of planar slices through that spatial distribution. What appear within the micrographs are sections of individual mitochondria, which have roughly elliptical shape. We use a maximum-likelihood scheme to infer the distribution of spheroid shapes from the |
|
|
observed distribution of elliptical shapes found within the micrographs.
Danny Flynn was supported as a trainee by NIH RoadMap Initiative award T90 DK07015. |
| |
|
|
Application of Machine Learning and Statistical Techniques to Study HIV Protease Inhibitors
By Rajni Garg, Srinivas Reddy Alla, Sunil Kumar, G N Sastry |
| HIV (Human Immunodeficiency Virus) is the pathogenic retrovirus and causative agent of AIDS or AIDS-related complex (ARC). HIV-1 protease (HIVPR) is one of the major viral targets for the development of new chemotherapeutics. Currently, many HIVPR inhibitor drugs are used in combination with other drugs. However, the use of current drugs regimens is compounded by several issues such as - adherence, tolerability, long-term toxicity, and drug- |
|
|
| and cross-resistance. Mutations also enable HIV to resist currently available treatments. Therefore, there is a growing need for the development of new chemotherapeutics with improved antiviral potency and favorable pharmacokinetic profile. This work presents a combinatorial QSAR (QSARomics) approach for the prediction of biological activity of cycloalkylpyranone HIV-1 protease inhibitors. A combination of linear (MLR, LMS, pace, and regression by descretization, additive) and non-linear (FFNN) regression models are developed to link the structures to their reported biological activity. Chemical descriptors (features) that encode hydrophobic, topological, geometrical and electronic features are calculated to represent the structures of the molecules. Genetic algorithm and principal component analysis were used for feature selection. This study is proposed to lead to the selection of the best computational model which may be used as a virtual screening tool to search pharmaceutical libraries for new leads. |
| |
|
|
Are Non-Covalent Interactions Related to Mutations in HIV-1 Proteases? A Data Mining Analysis
By Srinivas Reddy Alla, Rajni Garg, Xiaoyu Zhang, G N Sastry |
| AIDS and its related disorders, caused by retrovirus Human Immunodeficiency Virus Type 1 (HIV-1), are a major health concern worldwide. HIV-1 protease (HIVPR) is one of the major viral targets for the development of new chemotherapeutics. Currently, many HIVPR inhibitor drugs are used in combination with other drugs. However, the use of current drug regimens has several shortcomings. HIV-1 is also known to have several mutants. Therefore, the |
|
|
| development of new HIVPR inhibitors that are less toxic, more tolerable, convenient and active against drug-resistant (mutant) viruses is highly desirable. The study of non-covalent interactions (such as Cation-pi, pi-pi, C-H-pi, N-H-pi) in the biologically significant molecules has received lot of attention recently. The role of these weak interactions in biological recognition is well established. We have developed an in-house database and web-based tool to identify all the cation-aromatic interactions in a given protein. The developed tool along with the existing public domain servers viz. Capture, NCI server are used to evaluate non-covalent interactions in mutated HIV protease proteins. The data mining and evolutionary techniques have been applied to establish correlation between the mutation pattern in HIV-protease and their non-covalent interactions with the inhibitor (drug) molecule. This study will be helpful in the design of potent novel protease inhibitors with divergent mutational pathways. |
| |
|
|
In Search of a Practical Noninvasive Method for Contractility Measurements in Neonatal Rat Cardiocytes
By David Torres Barba, Paul Paolini |
| The objective of this project is to develop a convenient method for measuring contractile responses (shortening, or force, or some combination of these two parameters) of neonatal cardiocytes. This method can then be employed as a more quantitative characterization of pharmacologic effects of drugs on neonatal heart cells than has previously been available, and as is readily accomplished on adult myocytes. Methods of quantifying neonatal cell contraction as reported in the |
|
|
literature have required use of elaborate methods such as a proximity detector or an atomic force microscope (to measure the increase in cell elevation during contraction), and typically interfere with simultaneous optical recording of cell signals such as the calcium transient. The current project starts with digitizing optical images from a CCD camera mounted on a phase contrast microscope. The video images are later analyzed frame by frame using functions available in Matlab's Image Processing Toolbox. The software allows measurement of the cardiocyte's area, perimeter and optical density in each frame. Image optimization techniques such as segmentation, dilation, and edge detection are applied to a sequence of frames depicting a contracting cardiocyte. Changes in the cell's area can be used as a direct measurement of the cell's contractile response. Graphs of cell area and perimeter vs. time can be compared to and validated by graphs showing regional sarcomere shortening vs. time. Sarcomere length vs. time graphs are obtained from one or more "area of interest" (AOI) windows set by the operator to track sarcomere motion within small myofibrillar regions found scattered through the cell. Whole neonatal cell and regional sarcomere shortening curves demonstrate synchronous behavior, and can be compared with adult cardiocyte length vs. time curves.
David Torres Barba was supported as a trainee by NIH RoadMap Initiative award T90 DK07015. |
| |
|
|
Using GoogleRanks™ for Phage Phylogeny
By Annalinda Arroyo, Kristen Mecadon, Chrystian Irazoque, Rob Edwards, Forest Rohwer, Peter Salamon |
Using protein similarity, we define an approach to molecular based phylogeny for bacteriophages. The phylogeny uses phage-phage distances obtained by weighted averaging of the protein-protein distances between every pair of proteins. The weights in the averaging use GoogleRanks™ of the associated random walks. We expect to find an interesting and significant depiction of the relatedness between bacteriophage species.
|
|
|
| Kristen Mercadon was supported as a trainee by NIH RoadMap Initiative award T90 DK07015. |
| |
|
|
Cooperative Dynamics in Coupled Noisy Dynamical Systems Near a Critical Point: the dc-SQUID as a Case Study
By John L. Aven, Antonio Palacios, Patrick Longhini, Visarath In |
| The natural behavior of any process is that whenever it is affected by an external phenomenon that the outcome of the system is changed. Hence the output of a process can be changed by the input of the system. Many of these processes have the capability of either being in a steady state while at others times they may be kicked into a periodic or oscillating state. This type of behavior is common in a system of equations which is used to |
|
|
detect and measure an external phenomenon which will change the state of the system. Hence, theses systems of equations are generally referred to as sensors. Quite often a sensor's effectiveness is increased once it is in a region where it is oscillating. This, however, is not the only way to increase the effectiveness of a sensor. If one is to connect the sensors in one of many different ways, generally referred to as a coupling, the sensitivity may be increased. This work will hence be demonstrating the behavior of a particular type of sensor coupled in a few particular ways. The sensors which I deal with are known as dc-SQUID Magnetometers. The coupling schemes are nearest neighbor coupling schemes. From these coupling schemes we find that the response to an external magnetic field is greatly increased under simple circumstances. Magnetic sensors are common in everyday life. They are used in many different ways. They are found in geological equipment, biomedical equipment such as MRI machines, and applications to homeland defense such as mine detection.
John Aven was supported as a trainee by NIH RoadMap Initiative award T90 DK07015. |
| |
|
|
Shadow Structures to Standardize GoogleRanks™ between Disconnected Components
By Kristen Mecadon, Annalinda Arroyo, Karl Heinz Hoffman, Peter Salamon |
| Google™ Technology allows us to define ranks for states of a random walk. If the states are disconnected it creates a decomposable walk. The shadow graph can be used to make connections and compare the GoogleRanks™. The biological example of interest is the graph derived from the distance matrix on phage proteins with the shadow graph linking all proteins in the same phage. |
|
|
| Kristen Mercadon was supported as a trainee by NIH RoadMap Initiative award T90 DK07015. |
| |
|
|
Mobile Patient Health Care and Information System
By Prajakta Kulkarni, Yusuf Ozturk |
| Pervasive care and chronic disease management to reduce institutionalization is a priority for most western countries. The realization of next generation ubiquitous and complex pervasive health care systems will be a challenging task. This poster presentation outlines the requirements and design space of mobile pervasive health care systems. It also describes the architectural and design aspects of a prototype system built around these design constraints and provides test results to validate the concept. |
|
|
| |
|
|
A Metagenomic Examination of a Solar Saltern
By Beltran Rodriguez Brito, Linda Wegley, Matt Haynes, Mike Furlan, Kristen Marhaver, Jennifer Mueller, Steve Rayhawk, Ben Felts, Rob Edwards, Jim Nulton, Joseph Mahaffy, Peter Salamon, Francisco Rodriguez Valera, Forest Rohwer
Salterns are natural or man-made structures composed of several conjoined pools with progressively increasing concentrations of salt, used to extract several mineral salts from water. The biota of the pools changes as salinity rises, starting from a community resembling a common aquatic community and becoming halophile-dominates ecosystem. In the higher salinity ponds Archaea seems to be the most relevant domain, but the presence of extremely halophilic organisms from the Bacteria domain has also been reported. Here we report preliminary findings of our metagenomic studies of the population of several ponds of a solar saltern. |
| |
|
|
Effect of Competing Failures and Load Redistributions on Progressive Failure Predictability in Truss Structures
By Kun Marhadi, Satchi Venkataraman |
| The progressive failure of redundant truss structures was investigated. Energy absorption of the truss under progressive failure is taken as a measure of its performance. The non-deterministic energy absorption when small random design variations are introduced was calculated. Measures for robustness and predictability are proposed. Designs with different performance, robustness and predictability identified |
|
|
| using a deterministic optimization procedure are investigated and characteristics that contribute to better energy absorption, robustness and predictable progressive failure are identified. The presence of competing failures is shown to lead to competing failure sequences and the possibility of unpredictable failure. Investigating these sequences and the resulted energy absorption of those sequences indicates that energy absorption is a function of the stiffness of the competing member that fails first. To improve the total energy dissipated, failure sequences must be tailored so that the stiffer members whose failures do not lead to the structure becoming mechanism fails before the less loaded and less stiff members that are needed to keep the structure intact fail. To improve predictability, competing failures must be eliminated by separating the safety factors of critical members at each failure event. |
| |
|
|
Computational Modeling of a New Generation of High-sensitivity BioSensors
By Saravana Pitchaikani, Sam Kassegne, Kee Moon |
| PMN-PT (1-x)Pb(Mg1/3Nb2/3)O3-xPbTiO3 materials are used in various Micro system fields due to their High Piezoelectric coefficients. This Property makes them favorable for the use as biosensors PMN-PT is a perfect substitution for Quartz crystal. PMN-PT can detect rapid and minute resonant frequency changes in target. This Poster summarizes the computational modeling of a new generation high sensitivity single crystal PMN-PT chip as a BioSensor. |
|
|
| |
|
|
CFD & Experimental Results for MHD Micropumps for Lab-on-a-Chip System
By Amandeep Singh, Bao Nguyen, Sam Kassegne, Vaibhav Patel |
| Magneto Hydrodynamics (MHD) is a phenomenon related to interaction of the electrically conductive liquid and magnetic field. Micro pumps are increasingly being used based on the MHD physics for pumping biological and chemical specimens such as blood and saline buffers on Lab on the Chip devices or Micro Total Analysis System (TAS). This research reports analytical investigation of generalized 3-D MHD equations that govern the multiphysics of micro-pumps. |
|
|
| Based on the mathematical frame work developed for Computational Fluid Dynamics (CFD) and Non Dimensional numerical Analysis (NDA), the manuscript reports the additional effects of non uniform magnetic and electric fields, Joule Heating, pinch effect and M shaped flows in different flow channel geometries. |
| |
|
|
Statistical Cluster Analysis of Pharmaceutical Solvents
By Dong Xu, Nancy Redman-Furey |
| High efficiency in polymorph screening and crystallization optimization can be gained by judicious selection of solvents for the study design. Examination of all 57 (Class 2 and 3) pharmaceutical solvents may enable a complete study design but is costly in terms of time and resources. Based on a 17 descriptor dataset specifically constructed for all the Class 2 and Class 3 pharmaceutical solvents recognized by the International Conference of Harmonization (ICH), an optimal two- |
|
|
| stage cluster analysis was carried out together with principal component analysis as a dimensionality and colinearity reduction pre-processor. Both hierarchial average linkage cluster analysis and non-heirarchial K-means cluster analysis converged on a 20-cluster solution with strong statistical criteria support and excellent agreement in cluster membership, which can be reasonably interpreted from a chemical perspective. This 20-cluster solution is offered as an option for design of more efficient solid state screening studies. |
| |
|
|
Hyperons and Neutron Star Matter
By Parker Burrola, Fridolin Weber, Rodrigo Negreiros |
| The purpose of this project was to numerically solve the Equations of State for neutron star matter. We examined solutions corresponding to symmetric nucleon matter and also for asymmetric nucleon matter. I will also discuss the future aspects and goals of this project. |
|
|
| |
|
|
Ab Initio Study of the Torsional Dynamics of Aromatic Molecules
By Dong Xu and Andrew Cooksy |
| Accurate prediction of the torsional barrier height and potential energy surface of aromatic molecules such as tolane 1, 4-Bis (phenylethynyl) benzene, and Stilbene is achieved by MP2 calculations with spin-component-scaled correction and systematic extrapolation to the Dunning complete basis set limit as well as DFT methods with hybrid functionals. Benchmark tests based on the experimental data of benzene is used to determine the zero-point energy |
|
|
| correction. The torsional Schrodinger equation is solved numerically within the framework of Finite Element Method and the vibrational energy levels, wavefunctions, and other spectroscopic properties are obtained in excellent agreement with the experiments. This study provides a nearly complete computational solution to the torsional problems and may aid the exploration of similar in aromatic pi-conjugated system. |
| |
|
|
Combining In Vivo and In Silico Screening for Protein Stability
By Nesreen Barakat, Nora Barakat, John Love, Lisa Carmody |
| The correlation between protein structure and function is well established yet the role flexibility plays in protein function is currently being explored. Here we describe an in vivo screen in which the thermal stability and flexibility of a test protein is directly correlated to the transcriptional regulation of a reporter gene. The system entails the use of a chimeric protein contruct that consists of three covalently linked domains which includes a constant N-terminal DNA binding domain, a variable central test proteins, and a constant C-terminal |
|
|
| transcriptional activation domain. The ten test proteins are mutant variants of the ß1 domain of streptococcal protein-G that fairly evenly span a thermal stability range from as low as 38° C to greater than 100° C. When the chimeric construct contains a test variant of high conformational flexibility the reporter gene is up-regulated to a greater extent relative to more stable/less flexible variants. Mutant variants of the Gß1 domain were used to benchmark the screen and spectroscopic methods were employed to characterize the thermal and structural properties of each variant. The screen was also combined with in silico methods to interrogate a library of randomized variants for selection of mutants of greater structural integrity. |
| |
|
|
A convex-combination based algorithm for grid generation
By Elbano David Batista |
We propose an iterative technique for grid generation based on a convex combination of the nodes of the grid and the implementation of a grid-size function to control the dimension of the elements. The algorithm starts with an initial structured-mesh of the domain. Then, we update the position of the each node, p, by doing a convex combination of the points of the grid that define a polytope containing p.
The appropriate values for the weights of the combination are calculated by using a grid-size function, f, defined in the domain. f will not, necessarily, be equal |
|
|
to the dimension of the elements, but rather, the values of f will determine the relative distribution of the dimension of the elements in the grid. Thus, given the grid-size function f, the basic step of the algorithm are: generate an initial grid of the domain; calculate the weights associated to each node of the grid using f; and update node’s positions according to the combination mentioned before. The process ends when the highest difference between two consecutives grids is small enough.
The function f can be given by the user or estimated numerically according to certain considerations. This function allows us to have control on the refinement of the grid in any part of the domain. Also, its gradient is related with the orthogonality of the grid: as far as we maintain the gradient of f bounded, the orthogonality of the grid seems to be guaranteed. We tested the algorithm on several difficult problems, obtaining very smooth grids with high quality. |
| |
|
|
3D Inversion for Dynamic Rupture Parameters of the 2004 M6.0 Parkfield Earthquake, California
By Rosa Jimenez and Kim Olsen |
| 3D Inversion of the near-fault ground motion data for the 2004 M 6.0 Parkfield Earthquake in order to estimate a spatially-varying model of stress drop and slip-weakening distance using a slip-weakening friction law. The inversion is carried out by the Neighborhood Algorithm, a direct-search method without computing gradients. A 3D model of the crustal structure is included in the inversion. |
|
|
| |
|
|
