The Rowland NMR Toolkit (RNMRTK) is a software package for processing multidimensional NMR data and a platform for the facile implementation of novel NMR data processing methods. In addition to efficient and robust algorithms for traditional signal processing methods such as linear prediction extrapolation and discrete Fourier transformation, RNMRTK implements a very powerful and general algorithm for computing maximum entropy (MaxEnt) reconstruction, which is especially useful for processing non-uniformly sampled NMR data. RNMRTK provides rudimentary tools for analyzing and displaying spectra, but is best used in conjunction with more powerful and general-purpose analysis software. RNMRTK interoperates with nmrPipe, nmrDraw, XEASY, Sparky, NMRview, and other NMR software packages.


The predecessor to RNMRTK was VNMR, a program for off-line processing and graphical display of one-dimensional NMR data developed by Jeff Hoch at Harvard in the late 1970’s. Jeff began development of RNMRTK upon moving to the Rowland Institute for Science in 1981. RNMRTK V.1 was released in 1985 (Rowland Institute for Science Technical Memorandum #RIS-11t). V.2 was developed by Jeff Hoch and Alan Stern in the early 1990’s, in conjunction with writing the book “NMR Data Processing” (Wiley, 1996). RNMRTK was used to produce all the figures in the book. V.3 was a major overhaul of the architecture and code, mainly involving generalization of data structures to support up to 4-dimensional data.

The Rowland NMR Toolkit is a collection of programs, rather than a single monolithic program. The individual programs (tools) communicate with each other using the mechanism of shared memory. The first implementation of RNMRTK was developed on VAX computers from Digital Equipment Corporation running the VMS operating system. VMS provided the ability for programs to share memory and supported memory-mapped input/output via the demand pager (used to implement virtual memory). More recent implementations of RNMRTK exploit the shared memory support present in most Unix variants.


In addition to Jeff Hoch and Alan Stern, many others have contributed to RNMRTK, either through advice and counsel, or through code development.  These include David Rovnyak, Kuo-Bin Li, Peter Schmieder, Mehdi Mobli, Gerhard Wagner, John Skilling, Yingxi Lin, David Donoho, Frank Delaglio, Mark Maciejewski, David Ruben, Michael Gryk, Blagoje Filipovic, and Valeriy Gorbatyuk. Support from the National Institutes of Health (Grants GM-47467, GM-72000, and RR-201255) is also gratefully acknowledged.