Mass Spectrometry:  Remote Experimentation and Collaboration
University of Delaware
Overview
Laboratory
VNC-based solution
_____________Video-based descriptions___________
VNC-based solution
X meeting-based solution
 
Printable version
 

Project Goal. The project goal was to provide simultaneous, collaborative, remote control of a MALDI-TOF mass spectrometer located at the University of Delaware. A network-based solution eliminates the need for collaborators to travel to the laboratory in which the large, expensive equipment is housed. Researchers can instead mail their samples (analytes) to the lab. In the initial preparation, an analyte solution is mixed with a matrix solution and allowed to crystalize on a sample plate (probe). After the probe is inserted into the mass spectrometer, all aspects of the equipment control, data acquisition, and analysis can be performed at the instrument itself or at any number of remote locations at which collaborators work.

The control and analysis software is a collection of tightly coupled X Windows applications supplied by the instrument's vendor, Bruker Daltonics, and is run on an attached Sun ULTRA-5 workstation. Several approaches were examined to allow the X applications to be controlled and viewed by the remote collaborators. The preferred solution, discussed below, is currently used in joint research by faculty and students at George Washington University, Drexel University and the University of Delaware. The principal investigators are Professor Murray Johnston, Chemistry and Biochemistry Department at the University of Delaware, Professor Akos Vertes, Chemistry Department at George Washington University, and Professor Kevin Owens, Chemistry Department at Drexel University. Other investigators interested in similar use of this equipment should contact Professor Murray V. Johnston, Department of Chemistry and Biochemistry, University of Delaware.

Preferred implementation. Several commercial and non-commercial software approaches were implemented and tested. While no single approach was ideal, the AT&T Research Laboratory's VNC (Virtual Network Computing) software provided the most robust solution. It allowed researchers in multiple locations to take mouse and keyboard control at will. VNC provided a common view of the current windows displayed on the UD laboratory's screen. The researchers were not constrained to use a single workstation manufacturer or operating system. Testing was done on Sun (Solaris), SGI (IRIX), and PC (MS Windows) workstations.

Thanks to the generosity of AT&T Research, the VNC tools are open-source and in the public domain. The tools continue to be maintained by AT&T with contributions by the VNC user community.

Two less-successful, view-and-control-sharing approaches that were investigated were the following:

This approach did not require any changes to the Bruker software. The sharing is done at the X server level and several X servers can participate in the experiment. The particular implementation used was xmx, written by John Basik. The Bruker X clients that were started in the multiplexed XMX server could be shared by each remote participant's xmx client's virtual root window. The xmx software was designed and worked well in a homogeneous collection of workstations, and is heavily used that way in Sun workstation classrooms at Brown University. We encountered problems in our heterogeneous environment, largely associated with variations in color maps and fonts among various hardware and software environments.
The particular implementation used was VisualTek's xtvision. This approach required that the X meeting control program initiate all of the mass spectrometer's control and analysis programs. Because of the design of Bruker software's start-up Tkl/Tk scripts, this approach would not work well without a major rewriting of the Bruker start-up scripts.

IT project team in collaboration with
Research Data Management Services
Information Technologies / User Services
University of Delaware

Greg Forte
Dean Nairn
Dick Sacher
Anita Schwartz
Department of Chemistry and Biochemistry
University of Delaware

Rick Cox
Murray Johnston

Department of Chemistry
George Washington University
Akos Vertes

With support from NSF Grant No. 9808166


Problems or questions?
Printable Version
Copyright © University of Delaware, 2000.