Skip Navigation

Your Environment. Your Health.

I. Overview

NvAssign

Kirby, N.I., DeRose, E.F., London, R.E., Mueller, G.A. (2004) NvAssign: proteinNMR spectral assignment with NMRView. Bioinformatics, 7, 1201-1203

 

Overview:

Introduction

 

Nuclear Magnetic Resonance (NMR) protein studies rely on accurate assignment of resonances.

 

The general procedure is to:

 

  1. Pick peaks
  2. Cluster data from various experiments or spectra
  3. Assign peaks to sequence
  4. Verify assignments with spectra

 

Many algorithms already exist for automating the assignment process (step 3). What is lacking is a flexible interface so that a spectroscopist can easily move from clustering (step 2) to assignment algorithms (step 3) and back to verification of the algorithm output with spectral analysis (step 4).

 

The software module, NvAssign was written for use for NMRVIEW. The module provides a flexible interface to cluster data and interact with existing assignment algorithms. Further, the software module is able to read the results of the other algorithms so that the data can be easily verified by spectral analysis. This generalizability of the interface is demonstrated by connecting the clustered data with the assignment algorithms PACES and MONTE using previously assigned data for the protein pol lambda. The spectral analysis program NMRVIEW is able read in the output of these programs for simplified analysis and verification.

 

Software Improvements

 

This project has focused on interfacing the different steps of the NMR assignment process by altering the software package NMRView 5.0.4 (Johnson and Blevins, 1994). The previous assignment module of NMRView (CBCA.tcl) was improved upon to make the user’s experience more intuitive and fluid, and to implement other assignment algorithms. A side-bar indicator was added to let the user easily switch between the steps and be able to reference at a glance what step they are currently examining.

 

These steps in our terminology are defined by the side-bar buttons outlined below:

 

  1. Examine Peaks: the spectroscopist removes artifacts from the peak picking routines.
  2. Filter Peaks: Peaks from a 3D experiments are removed if they do not fall within set tolerances from a reference peaklist, usually the HSQC. (optional)
  3. Cluster Peaks:
    1. Link Intralist: the multiple peaks from the 3D experiments are linked to a single H, and N resonance.
    2. Link InterList: the linked peaks from the various above experiments are clustered into spin systems.
  4. Edit Clusters: The user can verify the linkages and clusters defined above.
  5. Match Clusters: At this point one can either
    1. Use the current NMRView MatchCBCA algorithm
    2. Output data to PACES, skip to VII
    3. Output data to MONTE, skip to VII
  6. Edit Matches: The connectivities defined by the HNCACB or similar experiments can be verified.
  7. Assemble Fragments: Connected spin systems, called fragments, are placed in the sequence of the protein where they can be verified

 

Addition improvements include several options that were added to the main window when executing functions that were previously buried in other hidden windows. Control of the frames containing visualizations of data was improved to easily change settings that are dealt with frequently in the process. The data is also now exportable to a format readable by the external packages PACES and MONTE to sequence groups of peaks. Further the results of these programs can be read into NMRVIEW so that the results can be verified by spectroscopic analysis. We have verified the utility of this approach by confirming our previous assignment results for the protein pol lambda (DeRose et al., 2003), using both PACES and MONTE.

Back to Top