--- /dev/null
+Installation:
+
+This package uses GNU autoconf generated configure scripts, see the
+file INSTALL.configure for generic configure documentation.
+
+Quickstart: Usually you'll just have to type:
+----------
+ ./configure
+ make
+and (as root)
+ make install
+
+Install Paths:
+-------------
+The main programs, library, man, and info pages are
+installed in the usual places by default
+(i.e. /usr/local/{bin,lib,man,info}). In addition the following
+install paths are used:
+header files /usr/local/include/ViennaRNA [includedir/ViennaRNA]
+parameter files /usr/local/share/ViennaRNA [datadir/ViennaRNA]
+extra scripts /usr/local/share/ViennaRNA/bin [datadir/ViennaRNA/bin]
+documentation /usr/local/share/ViennaRNA/doc [datadir/ViennaRNA/doc]
+
+All these locations can be changed by calling ./configure with
+appropriate options.
+
+If you like having all software packages separated in their own directories
+you could run e.g.:
+./configure --prefix=/opt/packages/ViennaRNA --datadir=/opt/packages/ViennaRNA
+resulting in /opt/packages/ViennaRNA/{bin,lib,man,info,include,doc} directories
+
+The installed location of the Perl module is determined by Perl's own
+configuration, independent of any configure options.
+To leave out the Perl modules run
+ ./configure --without-perl
+
+The programs AnalyseSeqs and AnalyseDists offer some cluster analysis
+tools (split decomposition, statistical geometry, neighbor joining,
+Ward's method) for sequences and distance data. To also build these
+programs add --with-cluster to your configure options.
+
+This release includes the RNAforester and Kinfold programs, which can also
+be obtained as independent packages. Running ./configure in the ViennaRNA
+directory will configure those two packages as well. However, for detailed
+information and compile time options, see the README and INSTALL files in
+the respective subdirectories.
+
+The Kinfold program can be used to simulate the folding dynamics of an
+RNA molecule, and is compiled by default. Use the --without-kinfold
+option to skip compiling Kinfold.
+
+The RNAforester program is used for comparing secondary structures using
+tree alignment. To compile RNAforester you'll need to first install the
+g2 graphic library, then run configure with the --with-forester option.
+
+Calculation of partition functions (via RNAfold -p) uses double
+precision floats by default, to avoid overflow errors on longer sequences.
+If your machine has little memory and you dont't plan to fold sequences
+over 1000 bases in length you can compile the package to do the computions
+in single precision by giving ./configure the --enable-floatpf option.
+
+Memory requirements for a sequence of length n will be approximately
+12*n^2 bytes for RNAfold -p in the double precision version, and
+ 4*n^2 bytes for the minimum free energy alone.
+
+
+The readseq program is not integrated in the build process. If desired,
+install readseq manually, following the instructions in the readseq
+directory.
+
+There are man pages for all executables, the library is documented in
+./doc/RNAlib-2.0.7.pdf as well as a set of HTML pages in
+./doc/html/index.html. Documentation in HTML can also be found at
+http://www.tbi.univie.ac.at/~ivo/RNA/
+
+
+System specific notes:
+
+MacOS X users:
+
+Depending on your version of gcc (< 4.4 ?) it might be necessary to deactivate OpenMP support by passing the option
+
+--disable-openmp
+
+to the ./configure script, since the OpenMP support is only partial in earlier versions of gcc shipped for MacOS X.
+
+Additionally, since MacOS X 10.5 the perl installation distributed with MacOS X seems to always include so called
+fat-binaries, i.e. binaries for multiple architecture types. Although this may be a nice portability feature of MacOS X
+it will break compilation of the perl interface to the RNAlib!
+To circumvent this problem you have two options
+
+1. Compile the perl interface without fat-binary support. Do so by setting the global shell variable ARCHFLAGS to
+include only the target architecture used for compiling RNAlib. E.g. do the following before executing make
+
+export ARCHFLAGS="-arch x86_64"
+
+2. Compile the complete ViennaRNA package with fat-binary support. Therefore, run the ./configure script in the
+following manner
+
+./configure CC="gcc -arch i386 -arch x86_64 -arch ppc -arch ppc64" \
+ CXX="g++ -arch i386 -arch x86_64 -arch ppc -arch ppc64" \
+ CPP="gcc -E" CXXCPP="g++ -E"
+
+The second option is not guaranteed to produce a working result, thus we suggest using the first one to create a
+single architecture build of the RNAlib perl interface!
+
+General remarks:
+
+Calculating the partition function for long sequences ( >400 bases)
+generates a lot of floating point underflows. On some system this can slow
+down the calculation dramatically because an exception is invoked for each
+underflow. How to avoid this problem is highly system specific:
+
+On SunOS 4.1 the remedy is calling nonstandard_arithmetic() add a -DSUN4 to
+the CFLAGS in lib/makefile to do this. Use CC=gcc since SUN's standard cc
+doesn't understand ANSI C.
+
+On HP machines running hpux-9 or hpux-10 add -DHP9 to the CFLAGS so that
+fpsetfastmode() will be called.
+
+
git://source.jalview.org / jabaws.git / blobdiff