Manpage of QUARTM
Section: Misc. Reference Manual Pages (l)
Return to Main Contents
QUARTM - quartett mapping to indentify the phylogenetic position of
a query sequence
quartm [options] < inputfile > outputfile
implements a version of the quartett mapping technique [K. Nieselt-Struwe
and A. von Haeseler, Mol. Biol. Evol 18: 1204-1219 (2001)] in
which a query sequence is compared to three or four groups of homologous
sequences in order to determine its phylogenetic position relative to these
groups. It uses Ken Clarkson's code for computing the convex hull.
ignore all quadruples that contain a '-' (gap character).
use frame <n> of coding nucleic acid sequences. Note that
there must not be frame-shifts in the input alignment!
use only the first four input blocks (see below).
verbose; printer a header specifying the options.
print a usage message and exit.
print version information and exit.
quartm uses a phylip-like alignment format in which each sequence
MUST stand on a single line preceeded by a name or ID of the sequence.
This name must not contain a whitespace (blank or tab character). The input
file contains either 4 or 5 blocks of data. The sequences in the first
block are used independently as query sequences. The data blocks are
separated by lines of the form
If four blocks are specified, the program uses blocks 2, 3, and 4 in
combination with all query sequences. In the case of five blocks the
query is compared to all six possibilities of combining two of the
four reference groups.
The program prints the list of the sequence IDs in each input group.
Then the mean support for each of the trees is listed together with
the width of the distribution of support values. A graphical representation
of the support values from each quadruple is plotted in Postscript.
Filenames are automatically generated from the IDs of the query sequences
and the reading frame, if applicable.
K. Nieselt-Struwe and A. von Haeseler, Mol. Biol. Evol
18: 1204-1219 (2001)
This man page documents version 0.2
Peter F Stadler.
The code is ugly because it consists of lots of distinct cases that are
just different enough to make it non-trivial not to deal with them
one-by-one. The program should probably (also) read a more standard
This document was created by
using the manual pages.
Time: 19:27:48 GMT, June 07, 2005