C++ OPPS, A New Software for the Interpretation of Protein Dynamics from Nuclear Magnetic Resonance Measurements

摘要

Nuclear magnetic resonance (NMR) is a powerful tool for elucidating protein dynamics because of the possibility to interpret nuclear spin relaxation properties in terms of microdynamic parameters. Magnetic relaxation times T1, T2, and NOE depend on dipolar and quadrupolar interactions, on chemical shift anisotropy and cross-correlation effects.Within the framework of given motional model, it is possible to express the NMR relaxation times as functions of spectral densities (Abragam, The Principles of Nuclear Magnetism; Oxford University Press: Clarendon, London, 1961), obtaining the connection between macroscopic observables and microscopic properties. In this context, recently Meirovitch et al. (Shapiro et al., Biochemistry 2002, 41, 6271, Meirovitch et al., J Phys Chem B 2006, 110, 20615, Meirovitch et al., J Phys Chem B 2007, 111, 12865) applied the dynamical model introduced by Polimeno and Freed (Polimeno and Freed, Adv Chem Phys 1993, 83, 89, Polimeno and Freed, J Phys Chem 1995, 99, 10995), known as the slowly relaxing local structure (SRLS) model, to the study of NMR data. The program C++OPPS (http://www.chimica.unipd.it/licc/), developed in our laboratory, implements the SRLS model in an user-friendly way with a graphical user interface (GUI), introduced to simplify the work to users who do not feel at ease with the complex mathematics of the model and the difficulties of command line based programs. The program is an evolution of the old FORTRAN 77 implementation COPPS (COupled Protein Probe Smoluchowski) and presents a number of new features: the presence of an easy to use GUI written in JAVA; high calculation performance thanks to features of C++ language, employment of BLAS (basic linear algebra subprograms) library (Blackford et al., Trans Math Soft 2002, 28, 135) in handling matrix-vector operations and parallelization of the code under the MPI (message passing interface) paradigm (Gropp et al., Parallel Comput 1996, 22, 789, Gropp and Lusk, User’s Guide for mpich, a Portable Implementation of MPI Mathematics and Computer Science Division; Argonne National Laboratory, 1996); possibility to predict the diffusion tensor of the protein via a hydrodynamic approach (Barone et al., J Comp Chem, in press). A cluster version of C++OPPS was also developed, which can be easily accessed by users via the web.