The Webbioc library was download from bioconductor http://bioconductor.org/

Info from Bioconductor
package: webbioc
Description: An integrated web interface for doing microarray analysis using several of the Bioconductor packages. It is intended to be deployed as a centralized bioinformatics resource for use by many users. (Currently only Affymetrix oligonucleotide analysis is supported.)
Version: 1.0.1
Author: Colin A. Smith 

Below is a brief description about each of cgi pages and how it was altered from
the original pages produced by Colin Smith.  It should be noted that the pages written by Colin S. are written in a very nice and concise manor
make heavy but efficient use of the Perl CGI module.

Overview:
Many of the pages were integrated into SBEAMS to facilitate easy data transfer
and storage of the results in a regularized directory tree. This should help the end user easily run
and analyze their data in a variety of ways.
Some of the scripts were altered quite a bit while others have been totally cut out, see below for more info.


Current Script Names:
affy.cgi           Batch.pm     FileManager.pm           multtest.cgi    Site.pm
annaffy.cgi        BioC.pm      job.cgi                           	 Upload_affy_get_expression_data.cgi
annaffysearch.cgi  cancel.cgi  Make_MEV_jws_files.cgi  Setup_Notes.txt   upload.cgi


### affy.cgi ###

This script was altered a bit to be able to efficiently talk with the upload script.  Tweaked some of the R code to process RMA and GCRMA
utilizing a different function from bioconductor.

### annaffy.cgi  ###
Script is no longer used.  All the annotation is pushed out into the analysis file

### annaffysearch.cgi  ###
Script is no longer used.

### Batch.pm  ### 
This module modified very little.  Tweaked the commands to the batch scheduler for wall time.

### BioC.pm  ###
This module was modified just a bit, mainly to deal with SBEAMS.  Added subroutine to deal with the sample groups XML file and tweaked methods
to make sure the files were routed properly within ISB.

### cancel.cgi  ###
Script no longer functioning.  This needs to be fixed.

### FileManager.pm  ###
The module was modified a bit to just become aware of what ISB file names look like.  This allows the script to automatically parse certain
bits of info from the name.

### job.cgi  ###
Removed some of the functionality of the original script, which allowed the user to enter their token key to hopefully retrieve a previous data
set.  All the data tracking is now done in the database.  Now this page just functions to make the temp html page to be displayed while
certain analysis functions are running.

### Make_MEV_jws_files.cgi  ###
This is a new script and allows the user to easily load data from SBEAMS into MEV via a java web start. The web start will load the normalized data file and
the sample groups the user has already defiend. See http://www.tigr.org/software/tm4/mev.html for more info on how to run MEV.  They have a ton of
documentation on how to use the program.

### multtest.cgi  ###
Script has been modified to allow the user to select between original multtest analysis choices or SAM analysis.  Wrote some R code to utilize the
siggenes library from bioconductor.  Both the multtest and SAM analysis will automatically loop through all the sample groups to produce unquie
conditions between a reference sample group and a additional groups for all the non-reference sample groups.  For example if four sample groups where added to the SAM
analysis page and A was designated as the reference sample (before the normalization step in the upload.cgi script) the following 3
condition would be setup.  A_vs_B, A_vs_C, A_vs_D.  Of course things will not loop if the F-test(ANOVA) is used.  The looping was setup to
setup the data to be loaded back into SBEAMS as a series of ratios.

### Site.pm  ###
A few new variables were added

### Upload_affy_get_expression_data.cgi  ###
This script is new and SBEAMS specific.  It takes the data from the differential analysis section and adds the data to the get expression
table.

### upload.cgi ###

This script was altered quite a bit and now serves as the main navigation point to reach the three different stopping points during 
analysis.  The three main points are File Groups, Normalized Data, Analysis Results

File groups, allows a user to group together CEL files from any number of projects and annotate what/why they have started this analysis
session.  The user no longer has to remember or keep track of a random key token provided in the original version.

Normalized Data, the user starting from a group of files can refine the association of which samples belong to which sample groups that will
be used for analysis.  In addition the user can add more sample groups and will select the so called reference sample group, more on this
later.  The actual choice of a normalization method is selected within the affy.cgi script.  Once the normalization run is complete the user
will have access to all the files generated during the run, including a tab delimited file of all the normalized gene expression values for all the
arrays entered into the normalization session.

Analysis Results: After a group of files are normalized a user can select to process the normalized output to find differentially expressed genes.  Now
they have two main paths with a variety of choices in each path.  One, they can view the data in MEV from tigr, see above for more info.  Two,
they can analyze they data within these cgi pages by selecting the multtest link, which now allows the user to analyze data with a variety of
t-test type algorithms original setup by Colin S. or to analyze the data via SAM, see above for more info.  After analysis is complete all
files are available for viewing.  The user can view a html of the "top" differentially expressed genes or have access to tab delimited files
of all the genes on the chip.  The user then can select to have all the data loaded back into SBEAMS into the GetExpression table, see above
for more info.







 

