Source Code

1. Downloading and compiling FASTML.
2. Running FASTML

Downloading and Compiling FASTML

Linux,Unix and other systems
Linux, Unix and other operating systems

1. Download the FASTML source code.
2. In order to unzip and untar the files please type:

unzip fastml.tar.zip
tar -xvf fastml.tar

This will create the following directories:

libs/phylogeny
programs/fastml

3. In some operating systems, you may use the makefiles to compile the program. If this does not work, skip to item 4.
Make sure you are in the directory where you unzipped the files, and type

cd libs/phylogeny

Type

make

in order to run the Makefile.
Now, type

cd ../../programs/fastml

to get to the fastml directory. Type

make

in order to run the Makefile.
This will result in an executable file called fastml which will reside in the programs/fastml directory.

4. In some systems (such as Unix), the makefiles will not be operable. Thus, follow steps 1-2 and compile directly using g++:
  • a. cd to the library where you unzipped the files.
  • b. Type

    mv libs/phylogeny/* programs/fastml/

  • c. cd to the fastml library

    cd programs/fastml

  • d. To compile, type
    g++ -O3 -o fastml *.cpp
This will result in an executable file called fastml which will reside in the programs/fastml directory.


If there are problems with the compilations (occasionally, with old version of g++) - please contact us and We'll try to help

Copyrights:
  • To modify the code, or use parts of it for other purposes, permission is requested. Please contact Tal Pupko
  • Please note that the use of the FASTML programs is for academic use only


Running FASTML
    FASTML uses flags in the command line arguments:

  • For help, type "fastml -h"
  • -s seq.aln = the sequence input file name.
  • -t tree.txt = the input tree file name (Newick format). If the user tree is not given, the program computes a NJ tree
    The program automatically recognize any of these sequence formats: Phylip, Molphy, Fasta, Mase, Clustal, or Nexus.

    • Here are two sample files:
     seq.aln (Fasta file format).
    tree.txt (Newick tree file format).
    You can try the program by typing:  'fastml -s seq.aln -t tree.txt' (without qoutes)


    The following models are supported:
    For Amino-Acids:
  • -mj JTT [default]
  • -ml LG
  • -md DAY
  • -mw WAG
  • -ma Jukes and Cantor (JC)for Amino Acid
  • -mr mtREV (for mitochondrial genomes)
  • -mc cpREV (for chloroplasts genomes)

    • For nucleotides:
  • -mn Jukes and Cantor (JC) for nucleotides

    • For Codons:
  • -my Yang et al. model
  • -me An empirical model


    • Results are printed for various output files:
  • tree file output in Newick format: tree.newick.txt (-x)
  • tree file output in ANCESTOR format: tree.ancestor.txt (-y)
  • joint sequences output file: seq.joint.txt (-j)
  • marginal sequences output file: seq.marginal.txt (-k)
  • joint probabilities output file: prob.joint.txt (-d)
  • marginal probabilities output file: prob.marginal.txt. (-e)
    • To change these default files name, use the flags: -x,-y,-j,-k,-d,-e (respectively) followed by the desired file name.

    Ancestral sequences output formats:
  • CLUSTAL (-qc) [default]
  • FASTA (-qf)
  • MOLPHY (-qm)
  • MASE (-qs)
  • PHLIYP (-qp)
  • Nexus (-qn)

    • Advanced options:
  • -a Treshold for computing again marginal probabilities [default: 0.9]
  • -b Do not optimize branch lengths on starting tree [by default branches and alpha are ML optimized from the data]
  • -c number of discrete Gamma categories for the gamma distribution [default: 8]
  • -f don't compute Joint reconstruction (good if the branch and bound algorithm takes too much time, and the goal is to compute the marginal reconstruction with Gamma).
  • -z The bound used. -zs - bound based on sum. -zm based on max. -zb both (default)
  • -p user alpha parameter of the gamma distribution (initial or fixed if -v is called)
  • -v alpha is fixed
  • -g Assume among site rate variation model (Gamma) For a joint-reconstruction with Gamma, the program implements a branch and bound algorithm (Pupko et al. 2002). Computing with the Gamma model is more accurate. However, this might be slow when the number of sequences is more than 30. If the alpha parameter of the Gamma distribution is known a priori, the -p flag is used to for input.
    If the alpha parameter is unknown, the program finds the ML estimate of alpha.

EXAMPLES:
    For example by typing:
  • 'fastml -s seq.aln -t tree.txt -mj -g' (without qoutes) - the program will compute ancestral reconstruction of the amino acid data in the seq.aln file according the tree file in tree.txt based on the JTT replacement model with among site rate variation

  • 'fastml -s seq.aln -t tree.txt' (without qoutes) - the program will compute ancestral reconstruction of the amino acid data in the seq.aln file according the tree file in tree.txt based on the default parameters



For any problems or questions please contact us at bioSequence@tauex.tau.ac.il.
Hope you enjoy!