 Overview.
 Program usage and options.
On the
following
page you will find a webbased user interface for the matrix program.
It allows you to set all relevant options as well as run the program on all
data.
Available options:
 sequence set to use
 You can either choose one of the sequence
sets used in [reference] or probe own sequences. To scan your own sequence
choose "Alternative sequence" from the pulldown menu and copy your
sequence in the provided input field. Note that the sequence must be in
fasta or multiple fasta format.
 matrix set to use
 The two provided matrices are ecoli,
the original E. coli matrix from Hertz' work [reference], and
mpneu, the M. pneumoniae matrix described in [reference]. You
can also choose "alternative matrix" and enter your own matrix in the
"alternative matrix" input field at the bottom of the page.
 GC content
 The matrix scores are defined as
log_{10}(f_{o}/f_{e}), where
f_{e} is the expected, and f_{o} the observed nucleotide
frequency. Of course, f_{e}
depends on the GC content. The GC value should therefore be adjusted to the
GC content of the probed sequences. In Mycoplasma pneumoniae, the GC
content of putative promoter regions (that is, 100 bases upstream of the
ATG) is about 36 %.
 use gap penalties
 Determines whether you want to use the gap
penalty. Gap penalties improve promoter detection. However, for the
computation of a novel matrix based on experimentally determined promoters
it is better to ignore the gap penalties.
 gap limits
 Set the allowed limits for the gap between the +1 and 10 regions, and
for the gap between the 10 and 35 regions.
 assume that the promoters are experimentally defined

Normally, the matrix tries to find the transcription start and +1 region as
well as the 10 and 35 regions. However, this makes no sense if the probed
sequences have experimentally defined transcription starts. If you check
this option, the matrix program will assume that the transcription starts
at the 11th base counting from the 3' end of the sequence
(sequence downstream from the transcription start is requiered for the
computation of a new matrix)
 use a treshold value
 Only putative promoters scoring above
the treshold will be shown and taken into account when computating a new
matrix.
 print out debugging information
 In case of any problems,
switching this option can help to pin down the cause.
 References.
