ISC'12

June 17-21, 2012

Hamburg, Germany

Contribution Details

 
Name: Modeling DNA Radiation Damage on Many-Core Architectures: The New GPU Version of SCELib for Microsoft HPC & Azure
 
Time: Wednesday, June 20, 2012
9:30 AM - 10:00 AM
 
Room:   Hall B
CCH - Congress Center Hamburg
 
Speakers:   Nico Sanna, CASPUR
 
Abstract:   There is a variety of processes initiated by the primary radiation impinging on DNA which can induce serious genetic effects, such as mutation [1]. At energies below the DNA ionization threshold the main mechanism responsible for single- and double-strand-breaks (SSBs and DSBs) has been linked to the action of low energy electrons (LEEs) [2-4] which have been proven to be the most abundant among the secondary species produced by the primary radiation [5]. The essential intermediates between the initial electron collision and the final molecular break-up have been identified as the Transient Negative Ions (TNIs), i.e. metastable states occurring in the scattering process which originate from the temporary trapping of the incident electron in the molecular potential field. The TNIs show a strong `local' character in the sense that they can essentially be associated to quantum resonances of the single DNA basic constituents [6, 7]. The characterization of these anionic intermediates is crucial in order to gather critical data for the analysis of the biological consequences of the electron interaction with the complex structures of the living tissues.
The approach we follow to model the electron-molecule interaction is based on a Single-Centre Expansion (SCE) of the molecular and incident electron wavefunctions about the centre of mass of the molecule and by solving the resulting scattering Volterra equations we calculate electron-molecule resonance positions and widths (further details on our theoretical procedure when applied to biological systems can be found in Ref. [8]). In our studies [9-15], these observables have been correlated with viable breaking pathways of various bio-systems such as the DNA molecular components (i.e., bases and (deoxy)-riboses) whose TNIs are possible precursors of SSBs and DSBs and the results used to directly model experimental data [11,12]. The whole set of codes we developed for electron-molecule scattering [16,17] has been implemented to efficiently run in parallel on most of the high performance computing architectures, and recently ported on many-core hybrid systems[18,19] by extending the method to larger molecular sections of the DNA. Nonetheless, the extremely large dimension of the molecular target to take into account require, with respect to present, resources with one (or more) order(s) of magnitude in computing power and a sort of computing “flexibility” where CPU power has to be coupled with the management of huge set of data. To this end, we are now releasing the new version of SCELib V4.0 with improved performance on multi and many-core (GPU based) computing systems, a package flexible enough to run even on cloud and with the ability to interact with large dataset on DBMS either on-premise or on cloud with Microsoft Azure.
We will present at the ISC12 conference our latest results on the application of SCELib and its accompanying codes to molecular systems as large as a DNA nucleotide and the resulting benchmarks of the new code with respect to the size of the molecular systems taken into account will be discussed in full detail.

References
[1] C. von Sonntag, The Chemical basis of Radiation Biology (Taylor and Francis, London, 1987).
[2] B. Boudaiffa, P. Cloutier, D. Hunting, M. A. Huels, and L. Sanche, Science 287, 1658 (2000).
[3] F. Martin, P. D. Burrow, Z. Cai, P. Cloutier, D. Hunting, and L. Sanche, Phys. Rev. Lett. 93, 068101 (2004).
[4] L. Sanche, Eur. Phys. J. D 35, 367 (2005).
[5] V. Cobut, Y. Fongillo, J. P. Patan, T. Goulet, M. J. Fraser, and J.-P. Jay-Gerin, Radiat. Phys. Chem. 51, 229 (1998).
[6] E.g. see, N. A. Richardson, S. S. Wesolowski, and H. F. S. III, J. Am. Chem. Soc. 124, 10163 (2002).
[7] E.g. see, N. A. Richardson, S. S. Wesolowski, and H. F. S. III, J. Phys. Chem. B 107, 848 (2003).
[8] I. Baccarelli, F. A. Gianturco, A. Grandi, R. R. Lucchese, and N. Sanna, Adv. Quantum Chem. 52, 189 (2007).
[9] A. Grandi, F. A. Gianturco and N. Sanna, Phys. Rev. Lett. 97, 018105 (2006).
[10] I. Baccarelli, A. Grandi, F. A. Gianturco, R. R. Lucchese and N. Sanna, J. Phys. Chem. B 110, 26240 (2006).
[11] I. Baccarelli, F. A. Gianturco, A. Grandi, R. R. Lucchese, N. Sanna, I. Bald, J. Kopyra and E. Illenberger, JACS 129, 6269 (2007).
[12] I. Baccarelli, F. A. Gianturco, A. Grandi and N. Sanna, Int. J. Quantum Chem. 108:11, 1878 (2008).
[13] I. Baccarelli, F. Sebastianelli, F. A. Gianturco and N. Sanna, European Physical Journal D (51:1), 2009, pp. 131--136.
[14] F. A. Gianturco, F. Sebastianelli, R. R. Lucchese, I. Baccarelli and N. Sanna, Journal of Chemical Physics (128:17), 2008.
[15] F. A. Gianturco, F. Sebastianelli, R. R. Lucchese, I. Baccarelli and N. Sanna, Journal of Chemical Physics (131:24), 2009.
[16] N. Sanna and F. A. Gianturco, Comp. Phys. Comm. 128, 139 (2000).
[17] N. Sanna and G. Morelli, Comp. Phys. Comm. 162, 51 (2004).
[18] N. Sanna, I. Baccarelli and G. Morelli, Comp. Phys. Comm. 180:12, 2544(2009).
[19] N. Sanna, I. Baccarelli and G. Morelli, Comp. Phys. Comm. 180:12, 2550(2009).
 
  • Tutorial Pass
  • HPC in Asia Workshop Pass
  • Conference Pass
  • Conference Pass or Exhibition Pass
    Satellite Event marked with * requires separate pass
  • Morning & Afternoon Coffee Breaks
    Midday Lunch Break
Program may be subject to changes.