The mfold package [16,17,18,19] consists
of a group of programs written in Fortran or C, and a group of Bourne
shell or Perl scripts. All of these programs and scripts run in a Unix
environment. The mfold software is currently running on a number of
different platforms, as indicated in Table 1.
|
In addition, some of the core programs of mfold have been translated into C++ and the resulting software is running on Windows 98/Windows NT (Intel hardware) with a point and click interface that uses proprietary Microsoft windowing tools. This program is called RNAstructure[20].
Up to version 2.2, the programs in mfold were interactive and
had to be run individually. Versions 2.3 and 3.0 provide a simple
command line interface, as described below, although it is still
possible to run individual programs interactively or, for that matter,
to write one's own scripts to link programs together in novel
ways. The RNA folding web server
(www.bioinfo.rpi.edu/~
zukerm/rna/form1.cgi)
provides an HTML interface to mfold that uses a similar, but
not identical script. This server will be described elsewhere. It's
main advantage is that anyone with a web browser can use it.
An early version of mfold (2.0) was ported to run on Mac computers. It lacked the energy dot plot and could only fold about 350 bases. We no longer recommend its further use, especially since the energy parameters it uses have not been updated. The current mfold version 3.0 runs in command line mode under Windows, but this implementation requires a type of Unix emulation program to be run.
Two Unix ``environment variables'' must be defined for mfold to function. The first, MFOLDBIN, defines the directory where all the executable files for mfold are stored. The second, MFOLDLIB, defines the directory containing all the free energy and other data files. Dynamic memory allocation is not yet available for mfold , and the largest fragment size that can be folded is defined by the value of the MAXN parameter in the ``Makefile'' file.
![]() | Michael Zuker Institute for Biomedical Computing Washington University in St. Louis 1998-12-05 |