These use cases describe how to run NUSSAMPLER jobs on the eNMR grid.
Gzipped tar files containing all the necessary files and an example of a job file for grid submission are provided for download. These use cases provide a basis for executing any NUSSAMPLER job on the eNMR grid.
Description of content:
This directory contains an example for generation of Non-Uniform Sampling (NUS) schedules, nD tables for running of sparse experiments on Varian or Bruker, with NUSSAMPLER
Created by Victor Jaravine / Anurag Bagaria / Peter Güntert / Henry Jonker, BMRZ, December 2008
Generation of NonUniform Sampling schedules: (file: usecase_nussampler.tgz)
This is an example of data generation with the program NUSSAMPLER.
Input file:
- demo3d.in : parameter file determining the generation of schedule
Output file:
- demo3d.hdr_3 : text file with the results
Execution on the eNMR grid: (commands are in italics)
1) Login as an eNMR grid user:
voms-proxy-init -voms enmr.eu
2) Submit job file:
glite-wms-job-submit -o nussampler.jobid -a nussampler.jdl
3) Check status:
glite-wms-job-status -i nussampler.jobid
4) When finished get back output:
glite-wms-job-output -i nussampler.jobid --dir nussampler-output
Comments :
1. This program only makes a sampling table. The resulting experimental data (after it is recorded according to the generated table) can be processed using MDD software or alternatively by a Maximum Entropy reconstruction software.
2. The NMR spectrometer software also include this program in the distribution (for Vnmr already since several releases and for TopSpin in its next release), with detailed instructions how to set parameters for matched acquisition and run sparse experiments. This program can work as a standalone program (i.e. on the eNMR GRID too). The previous versions performed only generation of 2D-4D tables, the current version supports 2D-nD case. Additionally, it performs checks for 'holes' in high-dimensional representation of the schedules (e.g. for 3D it is 23D representation) to avoid making of spectral artifacts due to random occurrence of too many missing FIDs in some rows or colomns of sampling grid in indirect dimensions of spectra, which is important during data processing. So, this software is not just a random number generator.