%
% NOTE -- ONLY EDIT THE .Rnw FILE!!!  The .tex file is
% likely to be overwritten.
%
%\VignetteIndexEntry{webbioc Demo Script}
%\VignetteDepends{}
%\VignetteKeywords{web, interface, demo}
%\VignettePackage{webbioc}
\documentclass[12pt]{article}

\usepackage{hyperref}

\textwidth=6.2in
\textheight=8.5in
%\parskip=.3cm
\oddsidemargin=.1in
\evensidemargin=.1in
\headheight=-.3in

\begin{document}
\title{Written Script for Demonstrating webbioc}
\author{Colin A. Smith}
\maketitle

\section*{Introduction}

This document is written for both new users and presenters to help
provide a scripted walk through of the Bioconductor web interface.
Because \verb+webbioc+ doesn't use R as the interface, this document
will not use any R code and will instead give written queues for
what to do in the web interface. For lack of straightforward and
public sample data, this document relies on users to provide their
own CEL files.

\section{Uploading CEL Files}

\begin{enumerate}

\item Go to the Upload Manager and start a new session by clicking
the \verb+Start New Session+ button.

\item Note the session token at the bottom of the page. It is very
important to record that token somewhere on the local computer so
that you can return to the files you upload. To make things easier,
click the \verb+Save Cookie+ button. That will store the upload
manager token in the browser for a week, making many of the tools
easier to use.

\item To upload files, click the top-left button to bring up a file
selection dialog box. Choose one of the CEL files you want to
process. Finally, click the upload file button.

\item Upload the rest of the CEL files that you want to process.
To make the other tools easier to use, upload the files in the
order you would like them listed. (The Upload Manager always lists
files in the order in which they were created/uploaded.) Always
upload files containing control samples before those containing
experimental samples. Again, this helps later on.

\end{enumerate}

\section{Preprocessing Affymetrix Data}

\begin{enumerate}

\item There are two ways to start preprocessing Affymetrix data:

\begin{itemize}

\item[-] Go to the affy page of the Bioconductor web interface. Enter
the Upload Manager token and number of CEL files and click the
\verb+Next Step+ button.

\item[-] From the Upload Manager, check all the CEL files you want 
to process. Click the \verb+affy+ button at the bottom of the page.

\end{itemize}

\item If you want to change the order of the CEL files, use the
popup menus in the File column. You may also specify a different
sample names if the filenames are not descriptive enough or are
too long.

\item For demonstration purposes, leave the processing method set
to RMA. It is the fastest method and produces good expression
measures. The custom methods can take from about 10 minutes to
almost a day. Leave the other checkboxes alone as well.

\item Click the \verb+Submit Job+ button to start the processing.
Your browser will be taken to a page with the job summary. It will
automatically refresh until your job finishes. A fast computer with
a reasonable number of CEL files shouldn't take any more than a
couple of minutes.

\end{enumerate}

\section{Differential Expression and Multiple Testing}

\begin{enumerate}

\item After the preprocessing finishes, click the \verb+multtest+
link and then click the \verb+Next Step+ button. (The upload manager
token should already be filled in for you.)

\item Select the ExpressionSet that you just created which will be at
the end of the list. Unless there are more than two types of
experimental samples (usually control and experimental), leave the
number of experimental classes as 2. Click the \verb+Next Step+
button.

\item As long as you followed previous recommendations and there
are an equal number of control and experimental chips, the web
interface should fill out the class label correctly for you.
Otherwise, give all the control samples a label of 0 and all the
experimental samples a label of 1. You may also chose to ignore
any samples you wish.

\item Select the statistical test you want to use to detect
differential expression. The first two t-tests tend to work well
but the choice is yours.

\item Select the multiple testing procedure you want to
use. The FDR methods are fairly effective when looking at the whole
chip at once.

\item Leave every other setting at its default value. However,
click the checkbox to copy the aafTable back to the upload manager.
We will use that in the next step. Click \verb+Submit Job+ to start
processing.

\item Once processing completes, look at the HTML file to see the
top 100 most significantly differentially expressed genes. Also
briefly look at the various plots produced.

The M vs. A plot shows log fold change (M) vs. log overall expression
(A). The red points show the same 100 genes in the HTML file.  The
Normal Q-Q plot compares the distribution of the t-statistics to
the normal distribution. The wings on either end show deviation
away from the normal distribution. The last plot shows how selective
the multiple testing procedure is over a range of error rates.

\end{enumerate}

\section{Annotation with Biological Metadata}

\begin{enumerate}

\item After you have finished looking at the diagnostic plots,
click the \verb+annaffy+ link. Scroll towards the bottom of the
page and click the \verb+Load File for Annotation+ button.

\item Select the aafTable that you just created which will be at
the end of the list. Click the \verb+Next Step+ button.

\item First select the type of chip that was used in the microarray
experiment.

\item Next select the aafTable and Data columns that you would like
to include in the annotated table. Mac users should use the Command
key to select multiple items. PC users should use the Control key.
Rich data sources with good linked information include Probe,
Description, LocusLink, Cytoband, PubMed, Gene Ontology, and Pathway.

\item Click the \verb+Submit Job+ and wait for the annotation to
complete. If no annotation appears in the resulting HTML file, go
back and check to make sure that you selected the correct chip.
From the resulting table, try clicking on some of the links to see
what sort of information they reveal. (Probe links require a free
registration with Affymetrix.)

\end{enumerate}

\section{Searching Biological Medadata}

\begin{enumerate}

\item Click the \verb+annaffy search+ link. First select the same
chip that you were using before. Then select \verb+Description+
for the metadata type.

\item If you are a presenter, solicit the audience for a class of
genes that you would like to search for. Try to select a word that
you expect will appear somewhat frequently in the gene descriptions.
Enter that single word in the text box.

\item Click the \verb+Search Text+ button to begin the search. Text
searches are not completely optimized and may take longer than you
expect. I fast computer will finish the text search in about 30
seconds.

\item Once the search completes, you can go to the HTML file to see
the descriptions that were found. The text file contains a comma
delimited list of just the probe ids that correspond to found genes.

\item If you wish you may copy and paste that list back into the
\verb+Test only these gene names+ box in \verb+multtest+ and thus
limit the scope of multiple testing.

\end{enumerate}

\end{document}