%
% NOTE -- ONLY EDIT Biobase.Rnw!!!
% Biobase.tex file will get overwritten.
%
%\VignetteIndexEntry{Iyer517}
%\VignetteDepends{Iyer517, Biobase}
%\VignetteKeywords{Expression Analysis, time course}
%\VignettePackage{Iyer517}


%
% NOTE -- ONLY EDIT THE .Rnw FILE!!!  The .tex file is
% likely to be overwritten.
%
\documentclass[12pt]{article}

\usepackage{amsmath,pstricks}
% With MikTeX 2.9, using pstricks also requires using auto-pst-pdf or running
% pdflatex will fail. Note that using auto-pst-pdf requires to set environment
% variable MIKTEX_ENABLEWRITE18=t on Windows, and to set shell_escape = t in
% texmf.cnf for Tex Live on Unix/Linux/Mac.
\usepackage{auto-pst-pdf}
\usepackage[authoryear,round]{natbib}
\usepackage{hyperref}


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

\newcommand{\scscst}{\scriptscriptstyle}
\newcommand{\scst}{\scriptstyle}

\bibliographystyle{plainnat} 
 
\begin{document}

\title{An {\tt exprSet} for the Iyer genomic time series database}
\author{VJ Carey, {\tt stvjc@channing.harvard.edu}}
\maketitle

\section{Overview}
Iyer, Eisen et al (Science 1999, v283 83-87)
report a cDNA-chip based experiment to illustrate
the transcriptional response of fibroblasts to serum.
The original data are archived in full at 
\url{genome-www.stanford.edu/serum/data.html}.
This package provides access to a subset of the
data leading to Figure 2 of their paper.
It would be worthwhile to provide high-level objects
representing the entire dataset, and this will be
taken up in the future.

\section{The {\tt Iyer517} {\tt exprSet}}
To get access to the data, install the
{\it Iyer517} package and then attach it:
<<>>=
library(Iyer517)
data(Iyer517)
@
A summary of the key dataset is:
<<>>=
show(Iyer517)
@
The first few expression records are:
<<>>=
exprs(Iyer517)[1:4,1:6]
@
Note that columns 1 to 13 correspond to sampling times
in the absence of cycloheximide (an inhibitor of protein synthesis)
and columns 14 to 19 correspond to sampling times
in the presence of cycloheximide.  The tags {\tt UNSYN}
and {\tt UNSYNC} are sampling from cells in exponential
replication.

\section{Replication of some findings}

\subsection{Figure 2}
To reproduce Figure 2 we need a color scheme and
some transformations.  The following seems
to do a reasonable job:

<<>>=
chg <- seq(.1,8,.01)
mycol <- rgb( chg/8, 1-chg/8, 0 )
CEX <- exprs(Iyer517)
CEX[CEX>8] <- 8
@
\begin{center}
<<fig=TRUE>>=
image(t(log10(CEX[517:1,1:13])),col=mycol,xlim=c(0,3),axes=FALSE,
 xlab="Hours post exposure to serum")
axis(1,at=(1:13)/13,lab=c("0",".25",".5","1","2","4","6","8","12","16","20","24","u"),cex=.3)
@
\end{center}
@

However, the time 0 column of Figure 2 in the paper
shows some variability.  This is hard to square with the
caption indicating that the data depicted are ratios relative
to time 0.

\subsection{The mean within-cluster trajectories}
To orient it seems we need clusters contiguous to
the boundaries of the image matrix, because there are gaps
of unspecified length between many of the clusters.

<<fig=TRUE>>=
par(mfrow=c(2,2))
plot(apply((CEX[1:100,1:13]),2,mean),main="Cluster A", log="y",
ylab="fold change", xlab="index in timing sequence")
plot(apply((CEX[101:242,1:13]),2,mean),main="Cluster B", log="y",
ylab="fold change", xlab="index in timing sequence")
plot(apply((CEX[483:499,1:13]),2,mean),main="Cluster I", log="y",
ylab="fold change", xlab="index in timing sequence")
plot(apply((CEX[500:517,1:13]),2,mean),main="Cluster J", log="y",
ylab="fold change", xlab="index in timing sequence")
@

The trajectories are very similar to those reported in the paper.
\section{Extended annotation}
An effort has been made to incorporate GO tags into this data resource.
<<>>=
data(IyerAnnotated)
print(IyerAnnotated[1:5,])
@
At the time of construction, at most 5 GO tags had been
associated with any probes in the dataset, and a large number
of probes lacked both Locus Link and GO tags.
\end{document}