Welcome to the CisOrtho Download Page.  This page is
intented for users who want to run CisOrtho or Cis on
their own computer.  Doing so allows a wider range of
parameter settings in all steps in the pipeline:

1.  initially isolating the regions to be searched (snip.plx)
2.  building the position weight matrix (hmmbuild from hmmer-2.3.2)
3.  running CisOrtho, using different parameters.


Input files consist of cisortho input:

e_snip
b_snip
eb_geneinfo
gene_names.txt
orthologs-2.00

and for hmmbuild input:

eb.null
eb.pri


To get started, you first need to create a DNA sequence alignment file
(say sites.txt) with the following format:

1  ATTGGCTTCGTTAA
2  ATTGGCTTAGTTTG
3  ATTGGTTTCGTTAA
4  ATTGGTTTAGTTAA
5  ATTAGCTTAGTTAA
6  ATTAGCTTAGTTTA
7  ATTAGCTTAGTTTA
8  ATTGGCCTAGTTAA
9  ATTGGCCTAGTTAA
10 ATTGGCTTCGTTTG
11 ATTGGCTTCGTTTA
12 ATTAGTTTCGTTAA
13 ATTGGTTTCGTTTA

(in which there are spaces, not tabs, between the numbers and DNA)

IMPORTANT:  The DNA sequences must be aligned.  Therefore, you need
to have two spaces after numbers '1' through '9'.  It is crucial that
the alignment is what you intend, since hmmbuild will happily produce
a position weight matrix from an unintended alignment.


Then, with files eb.pri, eb.null, and sites.txt in the current
working directory, (and hmmbuild.exe either in the current working
directory or in your path) run the command:

hmmbuild -F --prior eb.pri --null eb.null sites.hmm sites.txt

This will produce your position weight matrix file 'sites.hmm'.

(See the HMMER Userguide, from http://hmmer.wustl.edu for an explanation
of the effect of different values in eb.pri and eb.null, among other
options)



When you launch cisortho (either by clicking on the executable in Windows
or typing 'cisortho' at a unix shell prompt), you will see a page like this:

Welcome to CisOrtho.  <enter> refreshes the current
option values.  Enter option values by typing, e.g.:

        -p "mygene.pwm" -t 10000 -d 3<enter>


If an option value has spaces,
put quotes around the value, as in:

        -e "Elegans DNA Sequence"


Filenames should be relative to the current directory
For example, if an input file is in the directory above, it should
be called "../FileName" or "..\Filename"

cisortho
-p ""                              hmmbuild file from HMMER package
-t 0                               number of total top hits to consider.
                                   Use 0 to allow every gene -d hits.
-d 0                               number of top hits per gene to consider
-r 0                               number of top-scoring hit-pairs to
                                   report
-e ""                              elegans non-genomic regions
-b ""                              briggsae non-genomic regions
-g ""                              file containing mapping of elegans gene
                                   names
-i ""                              file containing gene
                                   strand,bounds,chromosome info
-o ""                              contains elegans-to-briggsae ortholog
                                   mapping
-h ""                              prefix for output HTML tables
-l ""                              directory for output HTML tables
(Type ? for Help)
->


At this point type:

-p sites.hmm<enter>
-t 10000<enter>
-d 3<enter>
-r 100<enter>
-e e_snip<enter>
-b b_snip<enter>
-g gene_names.txt<enter>
-i eb_geneinfo<enter>
-o orthologs-2.00<enter>
-h MyResults<enter>
-l OutputDir<enter>

Hit <enter> again to see the updated settings.
CisOrtho will look for file names from the current working directory,
or an absolute pathname.  So, if the file orthologs-2.00 resides in the
directory above, you should provide the relative path ../orthologs-2.00,
etc.  Also, the -l option must be an existing directory (whose path is
given relative as well).

When all options are correct, hit '=', which runs the program.

Please see the UserGuide as well for interpreting the results.
Good Luck.

Henry Bigelow
January 2004