June 2010 - Allocations-admins - mailman3.cels.anl.gov

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 30 Jun '10

30 Jun '10

Hello, A change in allocation has been requested: Requester: roux (Benoit Roux) Project: kv Title: Studies of voltage-vgated potassium channel Description: Voltage-gated potassium (Kv) channels are membrane proteins that respond to changes in the transmembrane potential by altering their conformation to allow the passive conduction of K+ ions across the cell membrane. These channels are tetrameric proteins, in which each subunit comprises six transmembrane (TM) helical segments (S1-S6). The ion conduction pore is located at the center of the tetrameric structure and is formed by the S5 and S6 helices from the four subunits. The first four segments (S1-S4) in each subunit form a voltage-sensing domain (VSD) that is located in the lipid membrane at the periphery of the central pore domain (PD). Upon depolarization of the membrane, the VSD in each subunit undergoes a voltage-dependent transition from a resting to an active conformation, which then leads the opening of the intracellular gate of the ion conduction pore. The conformational changes associated with the activation of Kv channels result in the transfer of an electric charge Delta-Q across the membrane, that can be measured experimentally as a small transient capacitive current. In the Shaker K+ channel, the "gating charge" corresponds to the transfer of 12-14 elementary charge (e) along the transmembrane electric field. Correspondingly, a change V in the membrane potential shifts the relative free energy of the closed and open conformations by V*Delta-Q. The gating charge DeltaQ is, thus, a key concept to explain how channel activation is coupled to the membrane potential. Initially postulated by Hodgkin and Huxley in 1952, it was first detected and measured more than 20 years later by Armstrong and Bezanilla. Ultimately, explaining the voltage-gating mechanism of Kv channels in molecular terms requires gaining knowledge of the active and resting conformations, and then showing how those conformations are able to account for the experimentally observed gating charge Delta-Q. Our strategy is to first carefully construct atomic models of the active and resting states of Kv channels that are consistent with all available experimental information using protein structure prediction algorithms. We also have a novel method to compute the gating charge using all-atom MD. Pathak et al. have generated detailed atomic models of Kv1.2 in the open/active and closed/resting states using the Rosetta-Membrane structure prediction program. The model of the open/active state complements the information missing from the X-ray structure of the Kv1.2 channel used as a template, while the S1-S2, S2-S3, and S3-S4 loops in the voltage sensing domain were modeled de novo. These results offer a promising starting point to expand our understanding of voltage gating in Kv channels. Some critical issues must be addressed. Recently, the stability of those structural models in the complex dynamical environment of the lipid bilayer was ascertained using molecular dynamics (Khalili-Araghi et al, 2010) and the gating charge was calculated using a new theory of the membrane potential (Roux, 2008). Our long-term goal is to refine the atomic models of the closed/resting and open/activated states and test their ability to account for the experimentally observed gating charge using all-atom MD and explicit lipid membrane environment. In the coming year, we have 3 specific sub-projects: 1) We want to incorporate additional experimental constraints in the refinement of the resting down-state of the voltage sensor (Macpos et al, 2007). Ernesto Vargas (a graduate student at the University of Chicago) is carrying out this project. 2) We want to predict the effect of side-directed mutations on the channels, namely the effect of substitution along the transmembrane helix S4. David Medovoy (a graduate student at the University of Chicago) is carrying out this project. 3) We want we want to refine the 3d structure of the bacterial voltage-gated channel KvAP using electron paramagnetic resonance (EPR). We will use a new mmethod that has been develop by us (Sompornpisut et al, 2008) and tested on other channels (Vasquez et al, 2008). This work will be carried out by a visiting scholar (Rong Shen). We had originally requested 250K nodehours for the fiscal year, but only half of that, 125K was granted by the Allocation committee. These systems run NAMD on 25 cores at a rate of about 1 day/nanosecond of MD. For a total of 50 ns of simulation total for the 3 projects means that we need about 300,000 nodehours. REFERENCES Pathak, M.M., V. Yarov-Yarovoy, G. Agarwal, B. Roux, P. Barth, S. Kohout, F. Tombola, and E. Y. Isacoff. (2007). "Closing in on the Resting State of the Shaker K+ Channel." Neuron 56, 124-140. Campos, F.V., B. Chanda, B. Roux, and F. Bezanilla. (2007). "Two Atomic Constraints Unambiguously Position the S4 Segment Relative to S1 and S2 Segments in the Closed State of Shaker K Channel." Proceedings of the National Academy of Sciences of the United States of America 104, 7904-7909. Sompornpisut, P., B. Roux, and E. Perozo. 2008. Structural refinement of membrane proteins by restrained molecular dynamics and solvent accessibility data. Biophys 95:5349-5361. Vasquez, V., M. Sotomayor, D. M. Cortes, B. Roux, K. Schulten, and E. Perozo. 2008. Three-dimensional architecture of membrane-embedded MscS in the closed conformation. J Mol Biol 378:55-70. Pan, A.C., D. Sezer, and B. Roux. (2008). "Finding Transition Pathways Using the String Method with Swarms of Trajectories." J Phys Chem B 112, 3432-3440. Roux, B. 2008. The membrane potential and its representation by a constant electric field in computer simulations. Biophys J 95:4205-4216. Khalili-Araghi, F., V. Jogini, V. Yarov-Yarovoy, E. Tajkhorshid, B. Roux, and K. Schulten. 2010. Calculation of the gating charge for the Kv1.2 voltage-activated potassium channel. Biophys J 98:2189-2198. Current: undetermined amount Justification: We had originally requested 250K nodehours for the fiscal year, but only half of that, 125K was granted by the Allocation committee. These systems run NAMD on 25 cores at a rate of about 1 day/nanosecond of MD. For a total of 50 ns of simulation total for the 3 projects means that we need about 300,000 nodehours. For the resources, we expect to be using NAMD runs of up to 256 CPU. This amounts to: 256 CPU * 24 hours * 50 days = ~300,000 Fusion will be an excellent platform to establish and execute the string method algorithm. Requested: 300000 A specific reason has been given: This is a very active project with several graduate students involved. It is also a collaborative project with experimentalists at the University of Chicago (Perozo and Bezanilla). This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 30 Jun '10

30 Jun '10

Hello, A change in allocation has been requested: Requester: roux (Benoit Roux) Project: Drude Title: Development and validation of new polarizable force field for molecular dynamics Description: At the present time, MD simulations are not only limited by the computational resources but also by the physical correctness and accuracy of the underlying atomic models. Without a systematic approach to generate, test and validate the accuracy of the force fields, the value of MD simulations will always remain in question. The development of potential functions for computer simulations has a long history and is still a very active field. Essentially all of the MD simulations of complex molecular systems (biological or not) are currently carried out using fixed-charge non-polarizable force fields, even though there is a broad consensus that electronic polarization effect ought to be included. The important role of polarization is particularly obvious for self-assembling amphiphilic molecules, containing polar or charged groups as well as nonpolar moieties (see below). A number of research groups are working at developing new algorithm to treat polarization, eit her with inducible point dipoles, fluctuating charges, or classical Drude oscillators. The latter approach, which we have adopted, is the most advantageous due to its simplicity and computational efficiency. In this project, we will establish a protocol to automatically optimize the parameters of the molecules of interest using simulations. We will develop a global derivative-free optimization optimization search method to drive the parameter search relying on information from the large scale simulations. This procedure will involve a parallelized Hamiltonian-Replica exchange algorithm. The goal is to develop, optimize, and validate the functional form and parameters of a new force field accounting for induced polarization in molecular dynamics simulations must be for a range of biomolecular systems (lipid membranes, protein in solution, liquid phases, etc...). # computational methods The computational method involves ab initio quantum mechanics, as well as classical molecular dynamics simulations and free energy perturbation. The induced polarization is represented by introducing a small auxiliary charged particle attached to the nucleus by a spring. The pair forms a Drude oscillator. It can be shown that this construction accounts properly for the physics of induced polarization in molecular interactions. # scalability issues The new force field is developed to be implementable in a wide range of existing molecular dynamics codes. Currently, the model is supported by CHARMM, which has modest scalability. Our initial implementation of the methods in NAMD has allowed to run some benchmark systems with 75,000 water molecules with the polarizable force field. These NAMD simulatinos scale up to 1,000 cores without problems. # experiments planned We must simulate fully hydrated bilayer and monolayer membranes of DPPC, DPPE, DOPC and POPC and PS to optimized the force field for phospholipid molecules. To increase the sampling, we will probably run those simulations in parallel-parallel mode with replica-exchange MD simulations. The planned tests on the large solvated systems are absolutely necessary. We are also developing ion solvation models in liquid N-methylacetamide. # software requirements We use Gaussian, Qchem, CHARMM, NAMD. # sizes of calculations run on and/or planned for Jazz or other systems On single MD runs, we expect to use up to 64 cores in parallel. On replica-exchange MD runs, we could easily use up to 512 cores or more. Each of the simulations are expected to require about several weeks non-stop. # expected number of project members The number of members is expected to be 5-6, including Benoit Roux, Yun Luo and Albert Lau and Janamejaya Chowdhary Chris Rowley, from University of Chicago. Current: undetermined amount Justification: MD simulation of lipid membrane bilayers with the polarizable force field . The rate is 1.5 days/ns with 256 cores, for a total of 4,608 SUs for each system. We have to examine different sets of parameters to optimize the lipid force field. Each must be simulated for about 5-10 ns, which amounts to a total of 300,000 SUs. MD simulation of ion solvation with the polarizable force field in N-methylacetamide (NMA) and water for 3 alkali ion, 4 alkali earth, and 6 transition metals. The rate is 3 days/ns with 64 cores, for a total of 4,608 SUs for each system. There are 13 systems total, yielding 60,000 SUs. Ab initio calculations on NMA-Ion clusters using the program Gaussian. Here we mainly carry out geometry optimization with different level of theories: DFT hybrid, DFT pure, DFT+D, and the most computationally expensive, RI-MP2. We have to optimize about 80 clusters (different ions with different numbers of NMA and configuration around the ion) with each level. This amounts to a total of 60,000 SUs. Requested: 300000 A specific reason has been given: This is a very active project with 5 people currently involved. This is a key year for the development of the force field because we are almost done. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 30 Jun '10

30 Jun '10

Hello, A change in allocation has been requested: Requester: iddir (hakim iddir) Project: TIES Title: Tailored Interfaces for Energy Storage (TIES) Description: We propose to investigate the above mentioned properties (SEI formation, structure, Li transport, and reactions at the interface) on model systems using density functional theory (DFT) as implemented in the Vienna Ab Initio Simulation Package (VASP) currently available on JAZZ. We will particularly focus on the most currently used anode material (graphite) and its interface with Li2CO3, as well as Li2CO3 properties themselves. We will also investigate Si based materials, such as Si carbides, Si single crystal surfaces, and their interface with liquid electrolyte. In this case we will particularly focus on the electrochemical aspect of the modeling. The scale of these calculations exceeds small computer cluster capacity. The capabilities and already tested performance of JAZZ are extremely suitable for the type of calculations we propose. In particular, VASP, the package that we will be using in this study does not scale very well beyond 48 processors, precluding efficient use of short-time large-scale capabilities typical for general use computer centers. In fact, the scaling of the calculations is dramatically reduced with the number of processors beyond 48. Preliminary calculations for this work show that at least 750000 core hours would be necessary to complete the proposed work. Expected number of project members: 4 Current: undetermined amount Justification: Preliminary calculations ran on the PNNL machines, show that a typical open shell structure optimization run for a single Li2CO3 monomer adsorbed on an edge site of a four-layer graphite substrate (4x2x2 super-cell, 166 atoms, 512 bands), requires on average about 1000 node hours. The study of Li2CO3 growth on graphite alone (basal and edge faces) will require the investigation of about 40 different configurations [(Li2CO3)x, x=1,2,3,4]. The notoriously CPU-expensive NEB calculations (charged cells- slow convergence) will require about 2000 node hours to study the diffusion of Li-ion in Li2CO3. The study of the binding and frequency calculations of several adsorbents on both faces of the graphite support, as well as the studies on Si single crystal surfaces, and Si based materials and their interface with liquid electrolyte will require at least an equal amount of time 40000 node hours. A total of about 750000 core hours would be needed to perform these calculations. Requested: 150000 A specific reason has been given: We have made good progress on this project using the initial allocated time on Fusion. However, the systems under investigation seem to require more computer time than initially estimated. Both the systems size as well as the number of configurations that need to be investigated in this research project turned out to be larger than expected. For instance Li2CO3/Graphite interface structures with two different film thicknesses of Li2CO3 and different relaxation constraints on graphite with 406 and 335 atoms for the thick and thin films, respectively, required about 48000 core hours each so far, and will need about 20000 to 30000 hours to complete the structure optimization. The completion of the structure optimization for these two systems with two different configurations (4 large systems total) alone would require about 80000 to 120000 hours. The study of the binding and frequency calculations of several adsorbents on both faces of the graphite support will require at least 40000 hours. A total of about 150000 core hours would be needed to continue these calculations. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 29 Jun '10

29 Jun '10

Hello, A change in allocation has been requested: Requester: sinclair (Donald Sinclair) Project: Lattice-QCD Title: Lattice simulations of Conformal and Walking Technicolor. Description: We perform simulations to evaluate the functional integrals of QCD-like theories formulated on a discrete space-time lattice to enable determination of the non-perturbative aspects these theories. These include the properties of these theories at non-zero temperature including the scales of confinement and chiral symmetry breaking, and such zero temperature properties as spectra, decay constants and the running of the gauge coupling constant. We are particularly interested in those theories where the coupling constant evolves particularly slowly, since these are candidate 'Walking Technicolor' theories. Related to these are theories with an infrared fixed point. Of special interest is to differentiate these two different types of behaviour. We are performing simulations of QCD-like theories which are models for Walking or Conformal Technicolor. We have been studying theories which are essentially QCD but with colour-sextet rather than colour-triplet quarks. We hope to measure the running of the QCD coupling constant. For 2 or 3 flavours, 2-loop perturbation theory predicts an infrared fixed point. For 2 flavours, it is possible that a chiral condensate forms before this fixed point is reached. If so the fixed point is avoided, the theory is confining and chiral symmetry breaks spontaneously. However, there is a region where the coupling constant evolves very slowly. These are the properties required for a walking technicolor theory. Simulations we have performed so far at finite temperature suggest that this theory does indeed walk. This is an ongoing program at NERSC. We wish to run small-lattice simulations of the 3-flavour theory, which we expect to have conformal, rather than walking behaviour on Fusion, for comparison. If we are correct, the properties of the 2 and 3 flavour theories should look rather different. Our sextet quark codes are based on our earlier triplet quark codes and use the RHMC simulation method. The Rational Hybrid Monte Carlo (RHMC) is a stochastic molecular dynamics algorithm. The functional integral of QCD is written as a partition function of a classical field theory evolving in a fictitious time. The determinant of the Dirac operator raised to a fractional power is calculated by introducing bosonic fields (pseudofermions), and sandwiching this Dirac operator raised to minus said fractional power between them. This fractional power of the Dirac operator is approximated to machine accuracy by a rational approximation. After defining this this theory on a discrete space-time lattice, the inversions required by the partial-fraction expansion of the rational approximation are performed using Krylov space methods, in particular a multi-shift extension of the conjugate gradient algorithm. A global Metropolis Monte-Carlo accept/reject step applied at the end of each t rajectory removes discretization errors introduced by the numerical integration of these stochastic equations of motion. We parallelize the code by assigning a fixed number of adjacent lattice sites to each MPI task. Network bandwidth ultimately limits how small a chunk of the lattice can be assigned to each task. Benchmarks of this code on NERSC's Cray XT4 (Franklin) give 68 Gflops on 48 cores for a 12^3*4 lattice. This is a fortran code with the parallelization performed using MPI. Current: undetermined amount Justification: The CPU requirement is the estimated time required to perform finite temperature simulations of the 3-flavour theory on a 12^3*4 lattice, and is based on the CPU time which we used to perform our simulations on the same size lattice for 2 flavours on Franklin. This assumes that the code performance on Fusion and Franklin will be similar. We will also need to perform simulations on an 8^3*4 lattice, but expect this will be done during the break-in period. The reason we are proposing to perform this project on Fusion is that the other machines we have access to schedule jobs so as to discourage such small jobs as ours. Running these jobs on Fusion will allow us to run larger lattice simulations on these other platforms. We have benchmarked our 12^3*4 codes on Franklin and found the following performances on various numbers of CPUs 8 CPUs = 10 Gflops 12 CPUs = 17 Gflops 16 CPUs = 26 Gflops 24 CPUs = 41 Gflops 48 CPUs = 68 Gflops On 48 CPUs, the measured communication overhead is 47%, indicating that we are approaching the point where communications are becoming a bottleneck. Requested: 500000 A specific reason has been given: I am currently running low on resources and wish to try and complete the project on which I am working before the end of FY2010. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: Comp_Nanocatalysis
by accounts＠lcrc.anl.gov 04 Jun '10

04 Jun '10

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Julius Jellinek Applicant's institution: ANL Applicant's division: CSE Project Name: Comp_Nanocatalysis Project title: Theoretical/Computational Studies of the Fundamentals of One-Component and Alloy Metal Based Nanocatalysis Relevant to Alternative Sources of Energy Associated funding: BES/DOE Other Systems: NERSC, ~400000 core-hours Science: The central objective of this project is to characterize the chemical reactivity and catalytic functionality of one-component and alloy metal nanoparticles as a function of their size, structure and composition, as well as the material and morphology of their supports. The reactions considered will be those of relevance to alternative sources of energy. Project description: We will primarily use the NWChem package. Project URL: http://www.anl.gov/catalysis-science/index.html Requested allocation: 300000 Justification: The allocation request is based on estimates made at NERSC The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 01 Jun '10

01 Jun '10

Hello, A change in allocation has been requested: Requester: zapol (Peter Zapol) Project: nanocatalysis Title: Catalysis in nanostructured materials Description: Molecular dynamics studies using many-body potentials as implemented in DL_POLY computer code and DFT studies in VASP will be used to simulate support structures covered by hydroxyl groups. Radial distribution functions will be compared to available experiments. The code uses 32 to 48 PEs for about 50,000 atom calculations. The geometries from molecular dynamics simulations will be used in studies of catalytic activity. We will study reactions on supported metal subnanometer clusters such as alkane dehydrogenation. We will also continue investigations on using thermodynamics to modify shape of catalytic particles. Density functional studies will be performed using VASP and GAUSSIAN codes to determine reaction energies and barriers. The VASP calculations typically use 24-64 processors for 24 hours to 48 hours. Expected number of members in the project is three to six. Current: undetermined amount Justification: We have efficiently used our jazz allocations and had a good performance Requested: 150000 A specific reason has been given: The estimated amount of time that we have obtained for this project was not sufficient to finish the proposed calculations. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] [LCRC Accounts] Project Allocation Request
by John Valdes 01 Jun '10

01 Jun '10

This project has been granted 300K to date out of the 500K they initially asked for. This request is effectively for the remaining amount of time yet to be granted. OK to give them 200K? Note that our limit is 150K, so we could give them 150K now and defer the remaining 50K 'til the next meeting (on 7/1). I say we give them the whole 200K now. John On Fri, May 28, 2010 at 10:16:35AM -0500, accounts(a)lcrc.anl.gov wrote: > > A change in allocation has been requested: > > Requester: jgreeley (Jeffrey Greeley) > Project: DFT_corrosion [...] > Requested: 200000 > > A specific reason has been given: > The project is producing exciting results, but we have realized that a number of additional configurations (beyond what we originally anticipated) need to be considered to completely describe the system in detail.

1 0