From stevenowicki@nyu.edu  Thu Oct 27 11:58:08 2005
Received: from mx4.nyu.edu (MX4.NYU.EDU [128.122.109.24])
	by mx.cims.nyu.edu (8.12.10+Sun/8.12.10) with ESMTP id j9RFw6Q8001781
	for <shasha@cs.nyu.edu>; Thu, 27 Oct 2005 11:58:06 -0400 (EDT)
Received: from [128.122.133.26] (PL-NOWICKI.BIO.NYU.EDU [128.122.133.26])
	by mx4.nyu.edu (8.12.10/8.12.10) with ESMTP id j9RFw0sR027133
	for <shasha@cs.nyu.edu>; Thu, 27 Oct 2005 11:58:01 -0400 (EDT)
Message-ID: <4360F9B6.6010206@nyu.edu>
Date: Thu, 27 Oct 2005 12:00:54 -0400
From: Steve Nowicki <stevenowicki@nyu.edu>
User-Agent: Mozilla Thunderbird 1.0.6-1.1.fc4 (X11/20050720)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Dennis Shasha <shasha@cs.nyu.edu>
Subject: LaTeX Source from Lab Meeting Presentation
Content-Type: multipart/mixed;
 boundary="------------030004050804090609020508"
X-Scanned-By: MIMEDefang 2.49 on 128.122.80.78
X-Greylist: Default is to whitelist mail, not delayed by milter-greylist-2.0.1 (mx.cims.nyu.edu [128.122.80.78]); Thu, 27 Oct 2005 11:58:08 -0400 (EDT)
Status: R
Content-Length: 15734

This is a multi-part message in MIME format.
--------------030004050804090609020508
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Attached is the source code and Makefile I used to create and compile 
the presentation from today's lab meeting.  In order to compile it, 
you'll need to install the LaTeX package "prosper" (which implements the 
presentation-specific stuff) and comment out the \includegraphics 
commands (since I didn't send you the graphics).

Feel free to let me know if you have questions about it.

- Steve


--------------030004050804090609020508
Content-Type: text/x-tex;
 name="presentation.tex"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="presentation.tex"

\documentclass[total,final,slideColor,colorBG,pdf,darkblue]{prosper}

\usepackage{multirow}

\title{PII, Bootstrapping, and FTA Protocol}
\subtitle{A lab meeting in 3 parts}

\author{Steven D. Nowicki}
\institution{Plant Molecular Biology Laboratory\\New York University}

\begin{document}

  \maketitle


  %% PII CONTENT


  \begin{slide}{Overview}

    Topics:

    \begin{itemize}
    \item \textbf{Theory on the functional role of PII in nitrogen sensing in \textit{Arabidopsis}}
    \item \textit{Application of hierarchical clustering with multiscale bootstrap to the analysis of experimental conditions}
    \item \textit{Development of a rapid genotyping protocol using FTA Cards}
    \end{itemize}

  \end{slide}



  \begin{slide}{PII - Introduction}

    \begin{itemize}
    \item PII protein in bacteria and cyanobacteria is a nitrogen sensor
    \item In the presence of low Gln:$\alpha$-ketoglutarate ($\alpha$-KG) levels, adenylates glutamine synthetase (GS), activating it.
    \item Deadenylates GS in high Gln:$\alpha$-KG
    \end{itemize}

  \end{slide}

  \begin{slide}{PII Model in Bacteria}

    \includegraphics[width=4.5in]{PII-ECOLI.ps}

  \end{slide}


  \begin{slide}{PII - Introduction}

    \begin{itemize}
    \item Function of PII in \textit{Arabidopsis} not well characterized
    \item Functional studies to date have not been able to elicit structural or molecular phenotypes in PII knockouts.
    \item PII may only be a minor player in N sensing in Arabidopsis (Ferrario-M\'ery et al., 2005)
    \item Redundancies in the pathway make phenotypic analysis difficult
    \end{itemize}

  \end{slide}


  \begin{slide}{Microarray Analysis of PII Mutant}

    \begin{footnotesize}

      \begin{itemize}
	\setlength{\itemsep}{1pt}
      \item 94 genes in mutant line are misregulated in the presence of nitrogen and carbon as compared to wild type
      \item Created a gene network in Cytoscape
      \item Looked for overrepresented gene functions
      \item Found overrepresentation of genes involved in:
	\begin{itemize}
	\item Sulfate metabolism
	\item Amino acid biosynthesis
	\item Balance of asparagine in plants
	\end{itemize}
      \item All three overrepresented gene functions easily connected to N sensing
      \item Results encouraging
      \end{itemize}
    \end{footnotesize}
  \end{slide}


  \begin{slide}{\normalsize{Large-Scale Microarray Expression Analysis}}

    \begin{itemize}
      \footnotesize{
      \item Nearly 1400 whole genome microarray chips were
	normalized and gene co-expression correlation values were
	calculated
      \item 571 genes correlated with PII at a 0.50 level or higher were
	analyzed for over represented GO terms using BioMaps.
      \item Common terms of significance: 
      }
      \begin{itemize}
	\footnotesize{
	  \setlength{\itemsep}{-3pt}
	\item unannotated: 113
	\item mitochondrion: 137
	\item chloroplast: 42
	\item energy: 93
	\item amino acid biosynthesis: 50
	\item amino acid metabolism: 57
	}
      \end{itemize}
    \end{itemize}
  \end{slide}


  \begin{slide}{PII Knockouts}

    \begin{itemize}
    \item Two T-DNA insertion knockouts were obtained from Salk.
      \begin{itemize}
      \item SALK\_095650 (PII09) has an insertion in the PII promoter
      \item SALK\_021878 (PII02) has an insertion in an intron {\tiny(Knockout proven successful in Ferrario--M\'ery, et al.)}
      \end{itemize}
    \end{itemize}

    \includegraphics[width=\linewidth]{PII_T-DNA_map.eps}

  \end{slide}


  \begin{slide}{PII - Work Done So Far}

    \begin{itemize}
    \item PII knockouts were plated with Columbia ecotype in 8 different conditions of C and N
    \item No dramatic phenotype response was noted in the first replicates, but these experiments need to be repeated
    \item We also took 5 plates and put them in the dark after 7 days...
    \end{itemize}
  \end{slide}

  \begin{slide}{PII In the Dark}
    
    \begin{tabular}{p{0.45\linewidth}p{0.45\linewidth}}

      \includegraphics[width=\linewidth]{dark2-1.eps} &  \includegraphics[width=\linewidth]{dark5-1.eps}\\
      \textbf{C Only} & \textbf{C+N} \\
	     {\footnotesize PII shows markedly reduced etiolation compared to WT} & {\footnotesize Significant etiolation in both WT and PII}\\
	     \begin{tiny}
	       \begin{tabular}{|c|c|}
		 \hline
		 \multicolumn{2}{|c|}{Mean Hypocotyl Length}\\
		 \hline
		 WT & PII \\
		 \hline
		 1.04 cm $\pm$ 0.09 & 0.47 cm $\pm$ 0.07\\
		 \hline
		 \multicolumn{2}{|c|}{p=0.0017}\\
		 \hline
	       \end{tabular}
	     \end{tiny} &
	     \begin{tiny}
	       \begin{tabular}{|c|c|}
		 \hline
		 \multicolumn{2}{|c|}{Mean Hypocotyl Length}\\
		 \hline
		 WT & PII\\
		 \hline
		 1.27 cm $\pm$ 0.04 & 1.21 cm $\pm$ 0.08\\
		 \hline
		 \multicolumn{2}{|c|}{p=0.5006}\\
		 \hline
	       \end{tabular}
	     \end{tiny}
    \end{tabular}
    \tiny{\textit{WT and PII seedlings in other conditions did not etiolate.}}

  \end{slide}


  \begin{slide}{PII In the Dark}
    
    \begin{tabular}{p{0.45\linewidth}p{0.45\linewidth}}

      \includegraphics[width=\linewidth]{dark2-1_cartoon.eps} &  \includegraphics[width=\linewidth]{dark5-1_cartoon.eps}\\
      \textbf{C Only} & \textbf{C+N} \\
	     {\footnotesize PII shows markedly reduced etiolation compared to WT} & {\footnotesize Significant etiolation in both WT and PII}\\
	     \begin{tiny}
	       \begin{tabular}{|c|c|}
		 \hline
		 \multicolumn{2}{|c|}{Mean Hypocotyl Length}\\
		 \hline
		 WT & PII \\
		 \hline
		 1.04 cm $\pm$ 0.09 & 0.47 cm $\pm$ 0.07\\
		 \hline
		 \multicolumn{2}{|c|}{p=0.0017}\\
		 \hline
	       \end{tabular}
	     \end{tiny} &
	     \begin{tiny}
	       \begin{tabular}{|c|c|}
		 \hline
		 \multicolumn{2}{|c|}{Mean Hypocotyl Length}\\
		 \hline
		 WT & PII\\
		 \hline
		 1.27 cm $\pm$ 0.04 & 1.21 cm $\pm$ 0.08\\
		 \hline
		 \multicolumn{2}{|c|}{p=0.5006}\\
		 \hline
	       \end{tabular}
	     \end{tiny}
    \end{tabular}
    \tiny{\textit{WT and PII seedlings in other conditions did not etiolate.}}

  \end{slide}


  \begin{slide}{PII In the Dark}
    \begin{minipage}{0.40\linewidth}
      \setlength{\unitlength}{2000sp}%
      %
      \begingroup\makeatletter\ifx\SetFigFont\undefined%
      \gdef\SetFigFont#1#2#3#4#5{%
	\reset@font\fontsize{#1}{#2pt}%
	\fontfamily{#3}\fontseries{#4}\fontshape{#5}%
	\selectfont}%
      \fi\endgroup%
      \begin{picture}(2536,2850)(3334,-4486)
	\thinlines
	    {\color[rgb]{1,1,1}\put(3526,-1786){\line( 1,-1){1050}}
	      \put(4576,-2836){\line( 1, 1){1050}}
	    }%
	    {\color[rgb]{1,1,1}\put(4576,-2836){\vector( 0,-1){1425}}
	    }%
	    {\color[rgb]{1,1,1}\put(3451,-3361){\vector( 1,-1){900}}
	    }%
	    {\color[rgb]{1,1,1}\put(3376,-1786){\vector( 0,-1){1350}}
	    }%
	    \put(5701,-1786){\makebox(0,0)[lb]{\smash{{\SetFigFont{10}{16.8}{\familydefault}{\mddefault}{\updefault}{\color[rgb]{1,1,1}N}%
	    }}}}
	    \put(3376,-1786){\makebox(0,0)[lb]{\smash{{\SetFigFont{10}{16.8}{\familydefault}{\mddefault}{\updefault}{\color[rgb]{1,1,1}C}%
	    }}}}
	    \put(3376,-3361){\makebox(0,0)[lb]{\smash{{\SetFigFont{10}{16.8}{\familydefault}{\mddefault}{\updefault}{\color[rgb]{1,1,1}PII}%
	    }}}}
	    \put(4276,-4486){\makebox(0,0)[lb]{\smash{{\SetFigFont{10}{16.8}{\familydefault}{\mddefault}{\updefault}{\color[rgb]{1,1,1}etiolation}%
	    }}}}
      \end{picture}%
    \end{minipage}
    \begin{minipage}{0.55\linewidth}
      \begin{small}
	The preliminary data suggest that PII may play a role in
	regulating amino acid synthesis.  In the dark, with no other
	source of carbon and malfunctioning AA synthesis machinery,
	carbon skeletons are not properly recycled and no carbon source
	is available to the shoots.
      \end{small}
    \end{minipage}
  \end{slide}

  \begin{slide}{PII May Regulate GS}
    \includegraphics[width=0.75\linewidth]{PII_pathway.eps}
  \end{slide}

  \begin{slide}{PII May Regulate GS}
    \includegraphics[width=0.75\linewidth]{PII_pathway_KO.eps}
  \end{slide}

  \begin{slide}{Follow Up}
    \begin{itemize}
    \item Repeat the experiments
      \begin{itemize}
      \item In the dark
      \item In light
      \end{itemize}
    \item If PII does regulate GS, then try to phenocopy with a GS inhibitor (i.e., MSX) in WT
    \end{itemize}
  \end{slide}




  %% MULTISCALE BOOTSTRAP CONTENT 

  \begin{slide}{Overview}

    Topics:

    \begin{itemize}
    \item \textit{Theory on the functional role of PII in nitrogen sensing in \textit{Arabidopsis}}
    \item \textbf{Application of hierarchical clustering with multiscale bootstrap to the analysis of experimental conditions}
    \item \textit{Development of a rapid genotyping protocol using FTA Cards}
    \end{itemize}

  \end{slide}

  \overlays{2} {
    \begin{slide}{Multiscale Bootstrapping}
      \fromSlide{1} {
	\begin{itemize}
	\item Purpose: determine confidence levels for clusters
	\item Method: non-parametric resampling $n$ times (i.e., 1000)
	\end{itemize}
      }

      \fromSlide{2} {
	\includegraphics[height=3.5cm]{bootstrap_diagram.eps}
      }

    \end{slide}
  }

  \begin{slide}{Multiscale Bootstrapping}
    \begin{itemize}
    \item Multiscale? --- Given sample size of $i$, don't just make resamples of size $i$
      \begin{itemize}
      \item Make resamples from $0.5i \ldots 1.5i$, for example
      \item Do each scaled resample $n$ times
      \item Avoids problem of local optima
      \end{itemize}
    \end{itemize}
  \end{slide}


  \begin{slide}{Multiscale Bootstrapping}

    \begin{itemize}
    \item Local Optima - start with a guess, try to improve.  The data can cause us to get stuck in a less than ideal position.
    \end{itemize}

    \begin{center}
      \includegraphics{hillclimb.eps}
    \end{center}

  \end{slide}



  \begin{slide}{What's Significant?}

    \includegraphics[height=0.95\textheight]{lung.noboot.epsi}

  \end{slide}


  \begin{slide}{Multiscale Bootstrapping}

    \includegraphics[height=0.95\textheight]{lung.epsi}

  \end{slide}


  \begin{slide}{Clustering Experiments}
    \begin{itemize}
    \item Which experiments make genes of interest behave similarly as in my experiments?
    \item This does NOT mean all induced or all repressed
    \item Utility:
      \begin{tiny}
	\begin{itemize}
	\item Previously unconsidered experimental conditions
	\item Circular analysis might find new co-regulated genes
	\end{itemize}
      \end{tiny}
    \end{itemize}
  \end{slide}

  \overlays{3}{
    \begin{slide}{Cluster Dendrogram}
      \begin{itemize}
      \item Analysis of 124 genes co-regulated with NiR at a correlation of 0.5 or above, clustering on slides\\
      \item With 1364 slides, the dendrogram isn't terribly readable. 
      \end{itemize}
      \onlySlide*{2}{  \rput(5,2){\includegraphics[width=14.0cm]{boot.eps}}  }
      \onlySlide*{3}{%
	Clearly, we can't use a graphic visualization at this level.
      }
    \end{slide}
  }

  \begin{slide}{Results of Experiment Clustering}
    \begin{itemize}
    \item Many clusters have only two or three genes -- take the 10 largest clusters with p $\leq 0.01$
    \item Compare experimental conditions of each slide in the cluster
    \end{itemize}

    \begin{tiny}

      \begin{center}
	\begin{tabular}{|r|p{0.45\linewidth}|r|r|}
	  \hline
	  Cluster   & Experimental Conditions & No. of Slides & AU\\
	  \hline
	  94        & Riken-Goda Hormone Treatments & 67 & > 0.999\\
	  \hline
	  139       & Heat shock experiments  & 26 & > 0.999\\
	  \hline
	  145       & Multiple stress conditions (cold, drought, osmotic, oxidative, UV, etc.) & 217 & > 0.999\\
	  \hline
	  151       & Osmotic stress & 11 & > 0.999\\
	  \hline
	  118       & Multiple stress conditions, control slides & 59 & >0.999\\
	  \hline
	\end{tabular}
      \end{center}
    \end{tiny}

  \end{slide}

  \begin{slide}{Gene Analysis}
    \begin{itemize}
    \item Clustering on genes can tell a better story
    \item Searching VirtualPlant, get all genes involved in nitrogen assimilation
    \item Using data from all 1364 experiments, cluster with multiscale bootstrap
    \end{itemize}
  \end{slide}

  \begin{slide}{Gene Analysis - N Assimilation}
    \includegraphics[height=0.95\textheight]{boot.nitroass.eps}
  \end{slide}



  %% FTA CARD PROTOCOL


  \begin{slide}{Overview}

    Topics:

    \begin{itemize}
    \item \textit{Theory on the functional role of PII in nitrogen sensing in \textit{Arabidopsis}}
    \item \textit{Application of hierarchical clustering with multiscale bootstrap to the analysis of experimental conditions}
    \item \textbf{Development of a rapid genotyping protocol using FTA Cards}
    \end{itemize}

  \end{slide}

  \begin{slide}{FTA Protocol}

    \begin{itemize}
    \item Rapid Genotyping
    \item Archival Storage of Samples
    \item Semi-high throughput
    \end{itemize}

    \begin{flushright}
      \includegraphics[width=0.40\linewidth]{whatman_card.eps}
    \end{flushright}

  \end{slide}


  \begin{slide}{FTA Protocol}
    \begin{footnotesize}
      \begin{enumerate}
      \item Place tissue sample on FTA card, close lid, vigorously smoosh using pestle or similar
      \item Remove any tissue remaining on the cover and allow to dry
      \item Using 1.6mm hole punch, punch hole and collect in PCR tube
      \item Wash $\times 2$ with FTA Purification Reagent
      \item Wash $\times 2$ with Tris-EDTA
      \item Allow to air dry
      \item Use disc in usual PCR reaction
      \end{enumerate}
    \end{footnotesize}
  \end{slide}

  \begin{slide}{Semi-High Throughput Techniques}

    \begin{footnotesize}
      \begin{itemize}
      \item Attach pipette tip to hole punch to place discs directly into PCR tubes
      \item Puncture bottom of PCR tube with $\frac{5}{8}$ inch 25 G syringe needle
      \item Cut off the bottom 2 thirds of a PCR tube.  Wrap in parafilm and
	insert into the end of a vacuum tube
      \item After each wash, insert sample tube into the end of the vacuum apparatus to remove wash fluid
      \item After last wash, leave each PCR tube in the vacuum for 5 minutes or so to speed drying
      \end{itemize}
    \end{footnotesize}

  \end{slide}



  \begin{slide}{Results}

    \includegraphics[height=0.85\textheight]{pcr.eps}

  \end{slide}


  %% THANKS

  \begin{slide}{Thanks}
    Special thanks to

    \begin{itemize}
    \item Rodrigo Guti\'errez
    \item Manpreet Katari
    \item Miriam Gifford
    \item Suzan Runko
    \item Gloria Coruzzi
    \item Dennis Shasha
    \item and everyone else in the lab
    \end{itemize}

  \end{slide}

\end{document}

--------------030004050804090609020508
Content-Type: text/plain;
 name="Makefile"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="Makefile"

FILE_BASE=presentation

all: ${FILE_BASE}.tex
	latex ${FILE_BASE}
	latex ${FILE_BASE}
	dvips -Ppdf -G0 -t a4 ${FILE_BASE}.dvi
	ps2pdf -dPDFsettings=/prepress ${FILE_BASE}.ps

clean:
	rm -f *.dvi ${FILE_BASE}.pdf *.log *.out *.aux ${FILE_BASE}.ps

--------------030004050804090609020508--

