APPLICATIONS
- Solvatochromic effect for the denaturation and mutation processes in DNA
- High Gain Reach in Fusion by Relativistic Compression Ion Bunches in Radiation Pressure Acceleration Regime
- Regional weather prediction model (EMS) run for National Environmental Agency
- High Resolution Climate Scenarios Construction using RegCM Model
- Quantum-chemical description of influence of the R-groups on the formation of peptide bond
- Coupling of ODE & PDE models for parallel numerical simulation of malignant tumor growth with migration, proliferation and angiogenesis
Research Team:
J. Kereselidze Tbilisi State University
M. Kvaraia and Z. Pachulia Sokhumi State University
G. Mikuchadze Georgian Research and Educational Networking Association
Short description:
The influence of the environment on the proton transfer between nucleotide bases has crucial importance for denaturation and mutation processes in DNA. For quantitative description of these processes, activation (ΔE#) and reaction (ΔE) energies of the proton transfer as well as lactam-lactim (KTLL) and amino-imino (KTAI) tautomeric equilibrium constants by the quantum-chemical DFT method are calculated. It is shown that decrease in the environment polarity (Er) due to mixing of ethanol with water (solvatochromic effect) leads to a decrease in the activation energy of the proton transfer and to an increase of the mutation frequency (vm), and at the same time to the tendency of DNA to denaturation. Hence, energy and kinetic characteristics of the proton transfer may be used for quantitative estimation of a solvatochromic effect in DNA. The validity of the solvatochromic effect is confirmed by the bathochromic shift of the DNA absorption band in the UV spectrum.
Publications:
1. Jumber Kereselidze, Marine Kvaraia, Zurab Pachulia, George Mikuchadze, Solvatochromic Effect for the Denaturation and Mutation Processes in DNA: Computational Study.
Springer: High-Performance Computing Infrastructure for South East Europe's Research Communities Modeling and Optimization in Science and Technologies Volume 2, 2014, pp 109-115.
2. Kereselidze J., Kvaraia M. Mikuchadze G. Quantum-chemical Modeling of the Cyclic-Pentameric Mechanism for the proton transfer in Imidazole Derivatives. Reserch Journal of Chemical Sciences, 2015, vol. 5(4) 89-91.
3. Kereselidze J., Kvaraia M. and Mikuchadze G. Quantum-chemical Modeling of the stacking mechanism for the 1H-4H proton transfer in piridine derivatives. DFT Study, Journal of Physical and Theoretical Chemistry, 12 (3), 277-279, 2015.
Title: High Gain Reach in Fusion by Relativistic Compression Ion Bunches in Radiation Pressure Acceleration Regime
Research Team:
D. Garuchava and K. Sigua Institute of Physics of Tbilisi State University
G. Mikuchadze Georgian Research and Educational Networking Association
Short description:
Study of interaction of femto-second pulses of extremely high intensive laser beam (10E23 wt/cm2) with a superdense plasma targets for the:
- Investigation of possibilities of creation super heavy (Z>110) non-stabile elements.
- Creation of flying relativistic mirrors (v > 0.6c) to vary frequency of reflected (from the mirrors) radiation.
- Investigation of possibilities of creation of steady compressive stress on the spherical De-He3 plasma target (taking into account thermal pressure and radiation reaction forces caused by laser beam reflection) to fire a fusion reaction.
Publications:
D. Garuchava and K. Sigua, X-ray Generation by Extreme Intensity Laser Driven Flying Mirrors, Journal Physics of Plasmas (in print).
Research Team:
N. Kutaladze – National Environmental Agency
N. Dekanozishvili - National Environmental Agency
G. Mikuchadze - Georgian Research and Educational Networking Association
Short description:
The Weather Research and Forecast Environmental Modeling System (WRF EMS or only EMS), which incorporates dynamical cores from both the US National Center for Atmospheric Research (NCAR) Advanced Research WRF (ARW) and the US National Center for Environmental Predictions' (NCEP) non-hydrostatic mesoscale model (NMM) releases into a single end-to-end forecasting system, is the main weather forecasting tool in Georgian Hydromet. The model (EMS) processes Global Forecast System (GFS) data set for use as initial and boundary condition information for a local area WRF simulation. For the reason EMS calculates local area forecast on the 15km grid at first and after that (using the obtained data) more detail regional forecast are executed on the 5km grid.
Model is run in operational (every day) mode for 5 days weather forecast, which in turn implies availability of weather forecast results for the prognostic services not later 10:30. If take into account the fact that GFS data set are ready only for the 08:10 the full time for model’s run is quite small and requires high performance computing (HPC) hardware. The model has 3-phases: 1- GFS data preprocessing, 2 – model’s main calculations, 3 – results post-processing with the preparation of prognostic maps.
For the model run we use GRENA Grid infrastructure: 16 cores of Intel Xeon E5-2670 (2.6 GHz, 20M L3-Cache), approximately 32 CPUs x Hours per run.
Research Team:
N. Kutaladze – National Environmental Agency
L. Megrelidze - National Environmental Agency
G. Mikuchadze - Georgian Research and Educational Networking Association
Short description:
Climate change’s future approximation is quite important for the regional and local (Country) scale. The perfect is climate scenario and has higher resolution the better is planning of adaptation and mitigation actions for the region. The bases of the scenario are Global Climate Models (GCM) long range (100+ years) prediction. For the preparation of the high resolution scenarios GCM prediction has to be recalculated by the Regional Climate Models (RCM). For the reason we used the model RegCM v4.3 developed by International Centre for Theoretical Physics (ICTP) in Trieste, Italy. The model was adopted for the South Caucasus region and executed several runs for the various time periods and boundary conditions, summary 200 years on the 20 Km grid.
The model run is time and CPUs high consuming, which implies the requirement of HPC infrastructure. We used GRENA Grid infrastructure: Intel Xeon E5520 (2.27GHz, 8M L3-Cache), approximately 9000 CPUs x Hours.
Publications:
Georgia’s Third National Communication to UNFCCC (in preparation)
Research Team:
Jumber Kereselidze - Javakhishvili Tbilisi State University
Marine Kvaraia - Sukhumi State University
George Mikuchadze - Javkhishvili Tbilisi State University
Short description:
The description of influence of the R-group on the formation of the peptide bond by quantum-chemical method of density functional theory (DFT) is carried out. The criterion of probability of the formation of the peptide bond has been constructed. In particular is shown the propensity to formation of the peptide bond is increased as a result of 1) decrease of the CO and NH bond orders (PCO and PNH) and of activation energy (ΔE#), 2) increase of OH and CN bond orders (POH and PCN), 3) exothermic property of this reaction (ΔE<0).
Publications:
Computational Molecular Bioscience (Scientific Research an Academic Publisher) Volume 4, Number 2, June 2014
DOI: 10.4236/cmb.2014.42004 (http://dx.doi.org/10.4236/cmb.2014.42004)
Coupling of ODE & PDE models for parallel numerical simulation of malignant tumor growth with migration, proliferation and angiogenesis
Research Team:
Manana Lortkipanidze - Tbilisi State University
Rati Devidze - Tbilisi State University
Giorgi Ghughunishvili - Tbilisi State University
Georgi Kupunia - Tbilisi State University
Nino Ptskialadze - Tbilisi State University
Ana Tsiskarishvili - Tbilisi State University
Short description:
We have studied two-compartment ODE model of angiogenesis and angiogenic inhibition (S.Rlichelson and J.T. Leith) and model of tumor growth with migration and proliferation (A. V. Kolobov, A. A. Anashkina, V. V. Gubernov, A. A. Polezhaev).
These models were combined together, so that new model describes tumor growth taking into account carrying capacity of vascular system and neovascularization as well as spatial migration of malignant cells caused by insufficient amount of oxygen and/or nutrients or their proliferation otherwise.
We have discretized this new model and implemented it using methods of parallel computing. The program calculates concentration of tumor cells in the point taking into account proliferation migration and angiogenesis. Expansion of the cells described by the diffusion equation and the concentration on the different layers are calculated. Each of the layer corresponds to the beginner one at the different time from the beginning of reaction.