\documentclass[11pt]{article}
\usepackage{graphicx}
\usepackage{cite}
\usepackage{hyperref}
\pagestyle{myheadings}
\markright{microfluidicsModels08; Revised: 1 Sept 2010}

\setlength{\topmargin}{0in}
\setlength{\textheight}{8in}
\setlength{\textwidth}{6.5in}
\setlength{\oddsidemargin}{0in}
\setlength{\evensidemargin}{0in}

\def\rcode#1{\texttt{#1}}
\def\fref#1{\textbf{Figure~\ref{#1}}}
\def\tref#1{\textbf{Table~\ref{#1}}}
\def\sref#1{\textbf{Section~\ref{#1}}}

\title{Filling Out Table 1}
\author{Kevin R. Coombes}
\date{30 August 2010; REVISED: 1 September 2010}

\SweaveOpts{prefix.string=Figures/microfluidicsModels08,eps=FALSE}
<<options,echo=FALSE>>=
options(width=88)
options(SweaveHooks = list(fig = function() par(bg='white')),eps=FALSE)
@ 

<<makeFiguresDirectory,echo=FALSE>>=
if (!file.exists("Figures")) {
  dir.create("Figures")
}
@ 

\begin{document}
\maketitle
\tableofcontents

\section{Executive Summary}
\subsection{Introduction}
The main goal of the study is to identify changes in gene expression
in CLL that are associated with clinical outcome (including overall
survival and time-to-treatment).

\subsubsection{Aims/Objectives}
In this report, we want to see if there are any differences in
clinical parameters between the training datset and the valuidation
dataset. 

\subsection{(Statistical) Methods}
Exploratory data analyses and summaries.  Fisher's Exact test.
Two-sample t-tests.

\subsection{Results}

\begin{itemize}
\item Only the (categorical) white blood cell count shows a
  statistical difference between training and validation.
\end{itemize}

\subsection{Conclusion}

It is time to write the manuscript and submit it....

\section{Loading the Data}

In the previous report (\rcode{microfluidicsModels06.pdf}), we stored
all of the relevant data in a binary R file.  We begin by loading that
data.
<<load.data>>=
load("withmod02.rda")
ls()
@ 

\subsection{Functions}

We use these functions repeatedly in oprder to simplify the code to
sumamrize the different clinical parameters inn the training and
validation datasets.

<<foo.blark>>=
foo <- function(column) {
  train <- cardAB.clinical[, column]
  valid <- cardC.clinical[, column]
  list(train=train, valid=valid)
}

blark <- function(stuff) {
  cat("Train\n")
  print(summary(stuff$train))
  cat("\nValid\n")
  print(summary(stuff$valid))
  if (inherits(stuff$train, "factor")) {
    f1 <- data.frame(Overall=stuff$train, type="Train")
    f2 <- data.frame(Overall=stuff$valid, type="Valid")
    print(summary(rbind(f1, f2)))

    x <- c(stuff$train, stuff$valid)
    y <- rep(c("Train", "Valid"), 
             times=c(length(stuff$train), length(stuff$valid)))
    tab <- table(x, y)
    te <- fisher.test(tab)
  } else {
    cat("Overall\n")
    print(summary(c(stuff$train, stuff$valid)))
    te <- t.test(stuff$train, stuff$valid)
  }
  print(te)
}
@ 

\section{Summary Information}

\subsection{Age at Diagnosis}

<<age>>=
age <- foo("AgeAtDx")
blark(age)
@ 

\subsection{Gender}

<<sex>>=
sex <- foo("Sex")
blark(sex)
@ 

\subsection{Rai Stage}

<<rai>>=
rai <- foo("CatRAI")
blark(rai)
@ 

\subsection{White blood cell count}

<<wbc>>=
wbc <- foo("CatWBC")
blark(wbc)
@ 

\subsection{Platelets}

<<plt>>=
plt <- foo("Platelets")
blark(plt)
@ 

\subsection{Serum Beta-2 microglobulin}

<<b2m>>=
b2m <- foo("CatB2M")
blark(b2m)
@ 

\subsection{Serum lactate dehydrogenase}

<<ldh>>=
ldh <- foo("Serum.LDH")
blark(ldh)
@ 

After log transformation:
<<logldh>>=
logldh <- foo("LogLDH")
blark(logldh)
@ 

\subsection{Mutation Status}

<<mut>>=
mut <- foo("mutation.status")
blark(mut)
@ 

\subsection{Light chain subtype}

<<ltch>>=
ltch <- foo("Light.chain.subtype")
blark(ltch)
@ 

\subsection{Zap-70 Protein}

<<zap>>=
zap <- foo("Zap70Protein")
blark(zap)
@ 

\subsection{Matutes Score}

<<matu>>=
matu <- foo("Matutes")
blark(matu)
@ 

\subsection{CD38}

<<cd38>>=
cd38 <- foo("CatCD38")
blark(cd38)
@ 

\subsection{Cytogenetic complexity}

<<cyto>>=
cyto <- foo("CatCyto")
blark(cyto)
@ 

\subsection{Splenomegaly}

<<spl>>=
spl <- foo("Massive.Splenomegaly")
blark(spl)
@ 

\subsection{hemoglobin}

<<hem>>=
hem <- foo("Hemoglobin")
blark(hem)
@ 

\subsection{Prolymphocytes}

<<pro>>=
pro <- foo("Prolymphocytes")
blark(pro)
@ 

\subsection{Hypogammaglobulinemia}

<<hgg>>=
hgg <- foo("Hypogammaglobulinemia")
blark(hgg)
@ 

\section{Time-to-treatment}

First, we get median time-to-treatment in the training data.
<<>>=
library(survival)
@ 
<<train.t2t>>=
survfit(Surv(TimeDiagnosis2SigTreat, NumericSigTreatment) ~ 1,
        data=cardAB.clinical)
@ 

Now we do the asme thing for the validation dataset.
<<valid.t2t>>=
survfit(Surv(TimeDiagnosis2SigTreat, NumericSigTreatment) ~ 1,
        data=cardC.clinical)
@ 

Finally, we test whether there is any difference between the training
and validation datasets.
<<full>>=
full.clinical <- rbind(cardAB.clinical, cardC.clinical)
full.clinical$Use <- factor(rep(c("Train", "Valid"),
                                times=c(nrow(cardAB.clinical),
                                  nrow(cardC.clinical))))
@ 

<<surv>>=
model <- coxph(Surv(TimeDiagnosis2SigTreat, NumericSigTreatment) ~ Use,
               data=full.clinical)
summary(model)
@ 

\begin{figure}
<<fig=TRUE,echo=FALSE>>=
colset <- c("red", "blue")
plot( survfit(Surv(TimeDiagnosis2SigTreat, NumericSigTreatment) ~ Use,
              data=full.clinical), col=colset, lwd=2,
     xlab="Time (months)", ylab="Fraction untreated")
legend("topright", levels(full.clinical$Use), col=colset, lwd=2)
@ 
\caption{Kaplan-Meier plot of time from diagnosis to treatment in the
  training and validation subsets.}
\label{km}
\end{figure}

\section{Appendix}

<<save.fc>>=
save(full.clinical, file='full-clinical.rda')
@ 

This analysis was run in the following directory:
<<getwd>>=
getwd()
@


<<lib2,echo=F>>=
sessionInfo()
@ 

\end{document}