LONI-TeraGrid-DEISA Interoperabilty Project

From TeraGrid Wiki

Jump to: navigation, search

(A Science-Driven Project Using Advanced CyberInfrastructure funded by NSF via a HPCOPS award to LONI)


Contents

[edit]

Project Description

Context: Federated Grids have the potential to facilitate scientific advances that would not be possible otherwise. These advances arise from the advantages that federated resources provide, such as: (i) Aggregated higher throughput, (ii) Lower time-to-solution, (iii) Exploitation of opportunistic scheduling, and (iv) Larger peak-resource utilization, at otherwise unreachable scales. The ability to concurrently utilize federated resources requires Application-level Interoperability. Application-level interoperability can have different interpretations, but in this case it is defined as the ability to utilize multiple machines without refactoring for the specifics of resources.


Scientific Context: A significant problem in the treatment of HIV is the development of drug resistant mutant strains. In order to tailor treatment to individual patients, clinicians routinely genotype the virus infecting a patient. Knowledge-based decision support software built on existing clinical records and associated viral genotypic data is then used to aid drug selection. As part of the Virolab project, the aim of this research is to enhance the existing decision support rules by using predictive molecular dynamics simulations of the available drugs with their enzymatic targets (the HIV-1 protease and reverse transcriptase enzymes). Clinicians have to make decisions on treatment regimens on a timescale of approximately a week, necessitating that our simulations and the subsequent analysis are turned around within this time period. In order to obtain accurate results, ensembles of simulations that can use many thousands of processors are necessary. Coveney leads the Virtual Physiological Human (VPH) Initiative (as coordinator of the VPH NoE) and represents a recent successful award to the DEISA Virtual Community Support Initiative, on behalf of the VPH community.

The ContraCancrum project (an EU FP7 VPH-Initiative funded project) aims to develop a composite multilevel platform for simulating malignant tumour development and tumour and normal tissue response to therapeutic modalities and treatment schedules. In collaboration with the ContraCancrum project, molecular level simulation, will investigate the binding affinities of tyrosine kinase inhibitors to epidermal growth factor receptors, an attractive target for anticancer therapies. MD techniques can be used to study the interactions between inhibitors and receptors in atomistic detail, and to predict the effect of different receptors/mutations on inhibitor binding affinities using free energy calculations.

Project Goal: This is a year long Science-Driven Project Using Advanced CyberInfrastructure funded by NSF via a HPCOPS award to LONI (one of the TeraGrid Resource Providers), which aims to establish TeraGrid-DEISA Interoperabilty on a firm but extensible footing, and will begin starting 01 June 2009.

The high-level aim of this project is to enable scientific applications to utilise the federated capabilities of the TeraGrid, DEISA and LONI systems, to enhance the understanding of HIV-1 enzymes and epidermal growth factor receptors (EGFR) implicated in lung cancer -- important VPH science drivers. Specifically, the aim of this project is to use several Replica-based and Replica-Exchange simulations for HIV-1 & EGFR research, on multiple TeraGrid, LONI and DEISA resources, working concurrently towards the solution of a single problem instance -- the rapid computation of free-energies of binding with high-levels precision. The project will also work closely with ContraCancrum project researchers.

This project aims to facilitate the above goals using general-purpose, extensible and scalable methods while establishing the scientific advantages arising from this computational capabilty. In addition to scientific advances, this project will provide working implementations and tools that can be utilised by a broad range of applications to utilize resources and effectively scale-out on the TeraGrid, DEISA and LONI.


Click here for a detailed description

[edit]

Project Partners/People

  • LONI/LSU
    • Shantenu Jha
    • Joohyun Kim
    • Owain Kenway (TBC)
    • Hartmut Kaiser
    • Lukasz Lacinski
    • Honggao Liu
    • Ole Weidner
  • DEISA
    • Herman Lederer
    • Stefan Heinzel
    • Gavin Pringle (EPCC)
    • Paolo Malfetti
    • Morris Riedel
    • Jules Wolfrat
  • OMII-UK
    • Neil P Chue-Hong
  • PSC
    • Mike Levine
    • Shawn Brown
[edit]

Kickoff Meeting

TIME NAME TITLE
09:00 - 09:15 Dan Katz Introduction & The TeraGrid Perspective
09:15 - 10:00 Shantenu Jha Application-level Interoperability: Introduction and Perspectives
10:00 - 10:30 Peter Coveney Scientific Motivation & The VPH Perspective (TBD)
10:30 - 10:45 Break
10:45 - 11:00 Hermann Lederer The DEISA Perspective
11:00 - 11:15 Morris Riedel Interoperability Experiences between Infrastructures
11:15 - 12:00 Focussed Discussion (Katz Lead; Jha Scribe) Making it Happen: Deriving Science Requirements
12:00 - 12:30 Focussed Discussion (Jha Lead; Katz Scribe) Making it Happen: Bringing it together & Time-lines
[edit]

References

1. Adaptive Distributed Replica-Exchange Simulations, A. Luckow et al, Theme Issue of the Philosophical Transactions of the Royal Society A, volume 367, 1898 (2009). PDF

2. Developing Autonomic Distributed Scientific Applications, Grids Meet Autonomic Computing, workshop held in conjunction with ICAC Barcelona. PDF

3. Rapid and Accurate Prediction of Binding Free Energies for Saquinavir-Bound HIV-1 Proteases, I. Stoica, S. K. Sadiq, P. V. Coveney, , Journal of the American Chemical Society, 130, (8), 2639-2648, (2008). DOI: 1021/ja0779250

4. Real Science at the Petascale, R. Saksena et al Theme Issue of the Philosophical Transactions of the Royal Society A, volume 367, 1898 (2009).

5. Production level scientific simulation management on international federated grids, S. J. Zasada, B. G. Cheney, R. S. Saksena, J. L. Suter, P. V. Coveney, and J. W. Essex. In Proceedings of the 2nd TeraGrid Conference, 2007

6. AHE

7. Clinically Oriented Translational Cancer Multilevel Modelling (ContraCancrum)

Retrieved from "http://www.teragridforum.org/mediawiki/index.php?title=LONI-TeraGrid-DEISA_Interoperabilty_Project"
Personal tools