%\VignetteIndexEntry{IsoGeneGUI Vignette}
%\VignetteAuthor{Setia Pramana}
%\VignetteDepends{}
%\VignetteKeywords{Dose-response microarray data GUI}
%\VignettePackage{IsoGeneGUI}



\documentclass[a4paper]{article}
%\usepackage[OT1]{fontenc}
\usepackage{Sweave}
\usepackage{url}
\begin{document}

\title{IsoGeneGUI Package Vignette}
\author{Setia Pramana, Dan Lin, Ziv Shkedy}
\date{April 2, 2010}
\maketitle

\section{Introduction}
The IsoGene Graphical User Interface (IsoGeneGUI) is a user friendly interface of the IsoGene package which is aimed to 
perform analysis of dose-response studies in microarray experiments. 
The IsoGeneGUI is developed for the user with no or limited knowledge about R programming so he/she can perform 
the analysis of dose-response in microarray setting  easily. This GUI was developed using tcl/tk package.
The statistical methodologies (test statistics, etc.)  used in this package are discussed by Lin et.al (2007, 2008, 2010). 


\section{Usage}
To run the package 
<<open, eval=FALSE >>=
library(IsoGeneGUI)
IsoGeneGUI()
@


\section{Menus}

The main window of the IsoGene-UI is presented in Figure \ref{fig:Main}. The package has four main menus: File, Analysis, Plots and Help. In the middle of the main window 
there is an info box which provides information about the data (availability and summary) and 
the result summary of the last performed analysis.



\begin{figure}
\includegraphics[scale=0.5]{./images/main}
\caption{\label{fig:Main}The main window of the ISoGeneGUI.}
\end{figure}



\begin{enumerate}
\item  File:
       \begin{enumerate}
       \item Open Data
       		 \begin{enumerate}
         		 \item R workspace (*.RData files)
          		 \item Excel or text file (*.xls or *.txt files)
          	 \end{enumerate}
      \item  Show Data
       \item  Exit
 		\end{enumerate}
\item Analysis:  
 	\begin{enumerate}
    \item Set seed
    \item  Likelihood Ratio Test E2 Analysis
    \item Permutation Analysis
    \item Significant Analysis of Microarrays (SAM)
    	\begin{enumerate}
         \item  SAM Permutation
        \item SAM Analysis
         \end{enumerate}	
	\end{enumerate}
\item Plot:    
	\begin{enumerate}    
    \item  IsoPlot
   	\item Permutation Plot
    \item SAM Plot
    	\begin{enumerate}   
          \item  Plot of FDR vs. Delta
          \item  Plot of number of significant genes vs. Delta
          \item Plot of number of False Positive vs. Delta
        \end{enumerate}
   	\item  User defined scatter plot
	\end{enumerate}
\item Help: 
	\begin{enumerate}    
   	\item  IsoGene Help
    \item  IsoGeneGUI Help
    \item  About
  	\end{enumerate}
\end{enumerate}



\section{Reading the Data}
In the first step of the analysis, the data is uploaded to the package using the File menu.
The package can read data in R workspace file (*.RData), excel and text files.
\subsection{R workspace files}
 The format of the gene expression data should be a matrix or table where the columns are the arrays 
 and the rows are the gene names. The dose information should be a vector or table contain the dose 
 levels which corresponds to the arrays in the expression matrix/table.\newline
 The package provides an R workspace example dataset called \texttt{dopamine2}. For the dopamine2 data, 
 the dose levels are given by
\small
\begin{center}
\begin{verbatim}
> dose
 [1] 0.00 0.00 0.01 0.01 0.04 0.04 0.16 0.16 0.63 0.63
[11] 2.50 2.50 0.00 0.00 0.00 0.01 0.01 0.01 0.04 0.04
[21] 0.16 0.16 0.63 0.63 2.50 2.50
\end{verbatim}

\end{center}
\normalsize
while the expression matrix has the following structure:
\small
\begin{center}

\begin{verbatim}
> dopamine[1:5, 1:6]
             X1        X2         X3        X4        X5
201_at  2.579138  2.318749  2.496895  2.456772  2.479480
202_at  2.140561  2.061804  2.131749  2.107638  2.086722
203_at  6.988566  6.620562  5.764725  6.326178  7.020716
204_at 11.081855  9.974999 10.790689 10.702516 10.544664
205_at 12.104545 12.076975 11.989770 12.151120 12.118520
\end{verbatim}
\end{center}
\normalsize



Part of the \texttt{dopamine2} can be obtained in folder "exampleData" inside the package. 


The full example data in R workspace, text and excel files can be obtained from: \url{http://www.ibiostat.be/software/IsoGeneGUI/index.html}

\subsection{Open the data set}


In order to upload the R workspace we choose in the file menu in Figure~\ref{gui3}a the following sequence:

\begin{center}

\begin{verbatim}
File > Open dataset > R workspace
\end{verbatim}
\end{center}

The package will automatically refer to the example data  \texttt{dopamine2} if the option open R workspace is choosen.


Text files and Excel files can be uploaded to IsoGene-GUI as well using following sequence in the dialogue box:

\begin{center}
\begin{verbatim}
File > Open dataset > Excel or Text files
\end{verbatim}
\end{center}

Once the the data is uploaded, information about the data is presented automatically in the info box of the main window.


\section{Exploratory Data Analysis}
\label{explore}

To explore the expression of the genes, the IsoGeneGUI provides the submenu \textbf{IsoPlot} 
from \textbf{Plots} menu. This feature displays the data points, sample means at each dose and an 
isotonic regression line. 

There are three input options to draw the isotonic regression plot, using gene name(s), row number(s) or using a range of row numbers. There are also three check boxes:
 \begin{enumerate}	
 \item Dose as ordinal.
  This option will draw the plot and treat dose as ordinal scale. The default will plot with dose as a continuous variable.
 \item Show isotonic regression curve for both direction.
 The default plot just display the isotonic trend/curve which is more likely fit the data. By checking this option, the isotonic trend for both directions will be displayed.
 \item Show summary of the data.
 This option is to provide a short summary of each selected gene.
 \end{enumerate}	



\section{Data Analysis}
The IsoGene-GUI package provides three options for analysis: (a) analysis with the likelihood 
ratio test statistic (LRT) using its exact p-values, (2) resampling based analysis and (3) significance analysis of microarrays (SAM).

\subsection{LRT Using exact p-values}
\label{LRTAsymp}
To perform the exact LRT we choose the following sequence in the analysis menu:

\begin{verbatim}
Analysis > Likelihood Ratio Test (E2)
\end{verbatim}

Then the main dialog box for analysis based on the LRT using exact p-values is shown in Figure 
\ref{fig:Exact}.

\begin{figure}[!h]

\includegraphics[scale=0.5]{./images/Exact}
\caption{\label{fig:Exact}The main dialog box of Likelihood Ratio Test $E^{2}$ analysis.}
\end{figure}

Note that the users can choose to perform the analysis for all the genes on the array or on pre defined subset. 
In addition, in order to adjust for multiplicity, 
we need to select from the menu the adjustment method to be used and the overall error rate.


\subsection{Resampling Based Methods}

Resampling based analysis can be performed using the main menu for resampling method shown in 
Figure~\ref{fig:Perm}. We need to specify  the number of permutations, the test statistic(s) to be 
used and the multiplicity adjustment(s). Similar to the $E^2$ menu, 
here we can specify default graphical displays.


\begin{figure}

\includegraphics[scale=0.5]{./images/permutationMain}
\caption{\label{fig:Perm}The main dialog box for resampling-based monotone trend test.}
\end{figure}


\subsection{Significance Analysis of Microarrays (SAM)}

The \textbf{IsoGene-GUI} also provides testing for the dose-response relationship under order restricted 
alternatives using the Significance Analysis of Microarrays procedure (SAM). To perform this analysis 
there are two main steps: (1) calculating the SAM regularized test statistic using permutation and 
(2) the SAM analysis. These two steps are performed by 
two sub menus in the SAM menu.
Then in the first step several options of the use fudge factor are provided. 
User can specify  to use \textbf{no Fudge Factor}, \textbf{Automatic Fudge factor} 
(fudge factor will be calculated using the methods described in the SAM manual), 
or the \textbf{fudge factor based on a specific percentile}.

After the statistics are computed or loaded from a file, we now can perform the SAM analysis by following the sequence:

\begin{center}
\begin{verbatim}
Analysis > Significance Analysis of Microarrays > SAM Analysis
\end{verbatim}
\end{center}

After specifying the test statistic, the main dialog box of SAM analysis, 
which shows the plot of the observed versus the expected test statistics and other options, 
will be shown (see Figure \ref{fig:SAM}). 

\begin{figure}

\includegraphics[scale=0.5]{./images/SAMAnalysisWindow}
\caption{\label{fig:SAM}The main dialog box for SAM analysis.}
\end{figure}


Note that, in this plot we can change the delta value (automatically using delta slider or 
manually using delta input box) and the FDR level. 


\section{Saving IsoGeneGUI Outputs}
Each plot produced from the analysis in IsoGeneGUI can be copied into clipboard which can be 
easily pasted into MS.Office documents, such as Ms.Word. It can be done by clicking the 
\textbf{Copy to Clipboard} button located in the bottom of the graph and then paste it directly 
to the document (or \texttt{Ctrl-V}). Furthermore, the plots can also be saved into several 
image formats (*.ps, *,png, *.jpeg, *.bmp, *tiff). After the format is selected/highlighted 
from the list, the user can 
click \textbf{browse} button to specify the name of the image file and also the file location.

Note that results of all genes and lists of significant genes from the three analysis above can be
shown and also saved into an \texttt{R-workspace} and/or an excel file by clicking \textbf{Save} button .




\section{Complete users' manual}
Users can access the IsoGeneGUI documentation  online or by installing the IsoGeneGUI
package locally. Then the following code can be typed at R prompt:

<<help, eval=FALSE >>=
if (interactive()) 
    {
	browseURL("http://www.ibiostat.be/software/IsoGeneGUI/index.html")
    }
	
@
or alternatively:
<<help, eval=FALSE >>=			
if (interactive()) {
library(IsoGeneGUI)
 IsoGeneGUIHelp()
}
@

Users can download example dataset and a complete Users' Manual from the site as well.


\section*{References}

\begin{enumerate} 
	\item Lin, D., Shkedy, Z., Yekutieli, D., Burzykowski, T., G\"{o}hlmann, H., De Bondt, A., Perera, T., Geerts, T. and Bijnens, L. (2007) 
	Testing for trends in dose-response microarray experiments: A comparison of several testing 
	procedures, multiplicity and resamplingbased inference. 
	\textit{Statistical Applications in Genetics and Molecular Biology}, \textbf{6(1)}, Article 26.
 
	\item Lin, D., Shkedy, Z., Burzykowki, R., Ion,
	T., G\"{o}hlmann, H.W.H., De Bondt, A., Perera, T., Geerts, T. and
	Bijnens, L. (2008) An investigation on performance of Significance
	Analysis of Microarray (SAM) for the comparisons of several
	treatments with one control in the presence of small variance genes.
	\textit{Biometrical Journal}, Multiple Comparison Problem, Special Issue,
	\textbf{50(5)}, 801--823.

\item Lin, D., Shkedy, Z., Yekutieli, D., Amaratunga, D. and
Bijnens, L., editors. \textit{Modeling Dose-response Microarray Data in Early Drug Development Experiments Using R}. 
Springer, (to be published in 2010).


\end{enumerate}

\end{document}