summaryrefslogtreecommitdiff
path: root/Master/texmf-dist/doc/latex/beamer/examples/a-conference-talk/beamerexample-conference-talk.tex
blob: f5faef720e98863b7da9754401b4f594c4c59b6a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
% Copyright 2007 by Till Tantau
%
% This file may be distributed and/or modified
%
% 1. under the LaTeX Project Public License and/or
% 2. under the GNU Public License.
%
% See the file doc/licenses/LICENSE for more details.



\documentclass{beamer}

%
% DO NOT USE THIS FILE AS A TEMPLATE FOR YOUR OWN TALKS¡!!
%
% Use a file in the directory solutions instead.
% They are much better suited.
%


% Setup appearance:

\usetheme{Darmstadt}
\usefonttheme[onlylarge]{structurebold}
\setbeamerfont*{frametitle}{size=\normalsize,series=\bfseries}
\setbeamertemplate{navigation symbols}{}


% Standard packages

\usepackage[english]{babel}
\usepackage[latin1]{inputenc}
\usepackage{times}
\usepackage[T1]{fontenc}


% Setup TikZ

\usepackage{tikz}
\usetikzlibrary{arrows}
\tikzstyle{block}=[draw opacity=0.7,line width=1.4cm]


% Author, Title, etc.

\title[Block Partitioning and Perfect Phylogenies] 
{%
  On the Complexity of SNP Block Partitioning Under the Perfect
  Phylogeny Model%
}

\author[Gramm, Hartman, Nierhoff, Sharan, Tantau]
{
  Jens~Gramm\inst{1} \and
  Tzvika~Hartman\inst{2} \and
  Till~Nierhoff\inst{3} \and
  Roded~Sharan\inst{4} \and
  \textcolor{green!50!black}{Till~Tantau}\inst{5}
}

\institute[Tübingen and others]
{
  \inst{1}%
  Universität Tübingen, Germany
  \and
  \vskip-2mm
  \inst{2}%
  Bar-Ilan University, Ramat-Gan, Israel
  \and
  \vskip-2mm
  \inst{3}%
  International Computer Science Institute, Berkeley, USA
  \and
  \vskip-2mm
  \inst{4}%
  Tel-Aviv University, Israel
  \and
  \vskip-2mm
  \inst{5}%
  Universität zu Lübeck, Germany
}

\date[WABI 2006]
{Workshop on Algorithms in Bioinformatics, 2006}



% The main document

\begin{document}

\begin{frame}
  \titlepage
\end{frame}

\begin{frame}{Outline}
  \tableofcontents
\end{frame}


\section{Introduction}

\subsection{The Model and the Problem}

\begin{frame}{What is haplotyping and why is it important?}
  You hopefully know this after the previous three talks\dots
\end{frame}

\begin{frame}[t]{General formalization of haplotyping.}
  \begin{block}{Inputs}
    \begin{itemize}
    \item A \alert{genotype matrix} $G$.
    \item The \alert{rows} of the matrix are \alert{taxa / individuals}.
    \item The \alert{columns} of the matrix are \alert{SNP sites /
        characters}. 
    \end{itemize}
  \end{block}
  \begin{block}{Outputs}
    \begin{itemize}
    \item A \alert{haplotype matrix} $H$.
    \item Pairs of rows in $H$ \alert{explain} the rows of $G$.
    \item The haplotypes in $H$ are \alert{biologically plausible}. 
    \end{itemize}
  \end{block}
\end{frame}


\begin{frame}[t]{Our formalization of haplotyping.}
  \begin{block}{Inputs}
    \begin{itemize}
    \item A genotype matrix $G$.
    \item The rows of the matrix are individuals / taxa.
    \item The columns of the matrix are SNP sites / characters.
    \item<alert@1->
      The problem is directed: one haplotype is known.
    \item<alert@1->
      The input is biallelic: there are only two homozygous
      states (0 and 1) and one heterozygous state (2).
    \end{itemize}
  \end{block}
  \begin{block}{Outputs}
    \begin{itemize}
    \item A haplotype matrix $H$.
    \item Pairs of rows in $H$ explain the rows of $G$.
    \item<alert@1> The haplotypes in $H$ form a perfect phylogeny.
    \end{itemize}
  \end{block}
\end{frame}


\begin{frame}{We can do perfect phylogeny haplotyping efficiently, but
    \dots}
  \begin{enumerate}
  \item \alert{Data may be missing.}
    \begin{itemize}
    \item This makes the problem NP-complete \dots
    \item \dots even for very restricted cases.
    \end{itemize}
    \textcolor{green!50!black}{Solutions:}
    \begin{itemize}
    \item Additional assumption like the rich data hypothesis. 
    \end{itemize}
  \item \alert{No perfect phylogeny is possible.}
    \begin{itemize}
    \item This can be caused by chromosomal crossing-over effects.
    \item This can be caused by incorrect data.
    \item This can be caused by multiple mutations at the same sites.
    \end{itemize}
    \textcolor{green!50!black}{Solutions:}
    \begin{itemize}
    \item Look for phylogenetic networks.
    \item Correct data.
    \item<alert@1->
       Find blocks where a perfect phylogeny is possible.
    \end{itemize}
  \end{enumerate}
\end{frame}


\subsection{The Integrated Approach}

\begin{frame}{How blocks help in perfect phylogeny haplotyping.}
  \begin{enumerate}
  \item Partition the site set into overlapping contiguous blocks.
  \item Compute a perfect phylogeny for each block and combine them.
  \item Use dynamic programming for finding the partition.
  \end{enumerate}

  \begin{tikzpicture}
    \useasboundingbox (0,-1) rectangle (10,2);
    
    \draw[line width=2mm,dash pattern=on 1mm off 1mm]
      (0,1) -- (9.99,1) node[midway,above] {Genotype matrix}
      (0,0.6666) -- (9.99,0.6666)
      (0,0.3333) -- (9.99,0.3333)
      (0,0) -- (9.99,0) node[midway,below] {\only<1>{no perfect phylogeny}};

    \begin{scope}[xshift=-.5mm]
      \only<2->
      {
        \draw[red,block]            (0,.5)   -- (3,.5)
          node[midway,below] {perfect phylogeny};
      }
        
      \only<3->
      {
        \draw[green!50!black,block] (2.5,.5)   -- (7,.5)
          node[pos=0.6,below] {perfect phylogeny};
      }

      \only<4->
      {
        \draw[blue,block]           (6.5,.5) -- (10,.5)
          node[pos=0.6,below] {perfect phylogeny};
      }
    \end{scope}
  \end{tikzpicture}
\end{frame}

\begin{frame}{Objective of the integrated approach.}
  \begin{enumerate}
  \item Partition the site set into \alert{noncontiguous} blocks. 
  \item Compute a perfect phylogeny for each block and combine them. 
  \item<alert@1-> Compute partition while computing perfect
    phylogenies. 
  \end{enumerate}

  \begin{tikzpicture}
    \useasboundingbox (0,-1) rectangle (10,2);

    \draw[line width=2mm,dash pattern=on 1mm off 1mm]
      (0,1) -- (9.99,1) node[midway,above] {Genotype matrix}
      (0,0.6666) -- (9.99,0.6666)
      (0,0.3333) -- (9.99,0.3333)
      (0,0) -- (9.99,0) node[midway,below] {\only<1>{no perfect phylogeny}};

    \only<2->
    {
      \begin{scope}[xshift=-0.5mm]
        \draw[red,block] (0,.5)   -- (3,.5) 
          node[midway,below] {perfect phylogeny}
                         (8,.5) -- (9,.5);

        \draw[green!50!black,block]
          (3,.5)   -- (6,.5)
            node[pos=0.6,below] {perfect phylogeny}
          (6.4,.5)   -- (8,.5)
          (9,.5) -- (10,.5);

        \draw[blue,block] (6,.5) -- (6.4,.5)
          node[midway,below=5mm] {perfect phylogeny};
      \end{scope}
    }
  \end{tikzpicture}
\end{frame}


\begin{frame}{The formal computational problem.}
  We are interested in the computational complexity of \\
  \alert{the function \alert{$\chi_{\operatorname{PP}}$}}:
  \begin{itemize}
  \item It gets genotype matrices as input.
  \item It maps them to a number $k$.
  \item This number is minimal such that the sites can be
    covered by $k$ sets, each admitting a perfect phylogeny.
    \\
    (We call this a \alert{pp-partition}.)
  \end{itemize}
\end{frame}


\section{Bad News: Hardness Results}

\subsection{Hardness of PP-Partitioning of Haplotype Matrices}

\begin{frame}{Finding pp-partitions of haplotype matrices.}
  We start with a special case:
  \begin{itemize}
  \item The inputs $M$ are \alert{already haplotype matrices}.
  \item The inputs $M$ \alert{do not allow a perfect phylogeny}.
  \item What is $\chi_{\operatorname{PP}}(M)$?
  \end{itemize}
  \begin{example}
    \begin{columns}
      \column{.3\textwidth}
      $M\colon$
      \footnotesize
      \begin{tabular}{cccc}
        0 & 0 & 0 & 1 \\
        0 & 1 & 0 & 0 \\
        1 & 0 & 0 & 0 \\
        0 & 1 & 0 & 0 \\
        1 & 0 & 0 & 0 \\
        0 & 1 & 0 & 1 \\
        1 & 1 & 0 & 0 \\
        0 & 0 & 1 & 0 \\
        1 & 0 & 1 & 0
      \end{tabular}%
      \only<2>
      {%
        \begin{tikzpicture}
          \useasboundingbox (2.9,0);

          \draw [red, opacity=0.7,line width=1cm] (1.7 ,1.9) -- (1.7 ,-1.7);
          \draw [blue,opacity=0.7,line width=5mm] (0.85,1.9) -- (0.85,-1.7)
                                                  (2.55,1.9) -- (2.55,-1.7);
        \end{tikzpicture}
      }
      \column{.6\textwidth}
      \begin{overprint}
        \onslide<1>
        No perfect phylogeny is possible.
        
        \onslide<2>
        \textcolor{blue!70!bg}{Perfect phylogeny}
        
        \textcolor{red!70!bg}{Perfect phylogeny}
        
        $\chi_{\operatorname{PP}}(M) = 2$.
        
      \end{overprint}
    \end{columns}
  \end{example}
\end{frame}

\begin{frame}{Bad news about pp-partitions of haplotype matrices.}
  \begin{theorem}
    Finding \alert{optimal pp-partition of haplotype matrices}\\
    is equivalent to finding \alert{optimal graph colorings}.
  \end{theorem}

  \begin{proof}[Proof sketch for first direction]
    \begin{enumerate}
    \item Let $G$ be a graph.
    \item Build a matrix with a column for each vertex of $G$.
    \item For each edge of $G$ add four rows inducing\\the
      submatrix $\left(
        \begin{smallmatrix}
          0 & 0 \\
          0 & 1 \\
          1 & 0 \\
          1 & 1
        \end{smallmatrix}\right)$.
    \item The submatrix enforces that the columns lie in different
      perfect phylogenies. \qedhere  
    \end{enumerate}
  \end{proof}
\end{frame}

\begin{frame}{Implications for pp-partitions of haplotype matrices.}
  \begin{corollary}
    If $\chi_{\operatorname{PP}}(M) = 2$ for a haplotype matrix $M$,
    we can find an optimal pp-partition in polynomial time. 
  \end{corollary}

  \begin{corollary}
    Computing $\chi_{\operatorname{PP}}$ for haplotype matrices is
    \begin{itemize}
    \item $\operatorname{NP}$-hard,
    \item not fixed-parameter tractable, unless
      $\operatorname{P}=\operatorname{NP}$, 
    \item very hard to approximate.
    \end{itemize}
  \end{corollary}
\end{frame}


\subsection{Hardness of PP-Partitioning of Genotype Matrices}


\begin{frame}{Finding pp-partitions of genotype matrices.}
  Now comes the general case:
  \begin{itemize}
  \item The inputs $M$ are \alert{genotype matrices}.
  \item The inputs $M$ \alert{do not allow a perfect phylogeny}.
  \item What is $\chi_{\operatorname{PP}}(M)$?
  \end{itemize}
  \begin{example}
    \begin{columns}
      \column{.3\textwidth}
      $M\colon$
      \footnotesize
      \begin{tabular}{cccc}
        2 & 2 & 2 & 2 \\
        1 & 0 & 0 & 0 \\
        0 & 0 & 0 & 1 \\
        0 & 0 & 1 & 0 \\
        0 & 2 & 2 & 0 \\
        1 & 1 & 0 & 0 
      \end{tabular}%
      \only<2>
      {%
        \begin{tikzpicture}
          \useasboundingbox (2.9,0);
          
          \draw [red, opacity=0.7,line width=1cm] (1.7 ,1.3) -- (1.7 ,-1.1);
          \draw [blue,opacity=0.7,line width=5mm] (0.85,1.3) -- (0.85,-1.1)
                                                  (2.55,1.3) -- (2.55,-1.1);
        \end{tikzpicture}
      }
      \column{.6\textwidth}
      \begin{overprint}
        \onslide<1>
        No perfect phylogeny is possible.
        
        \onslide<2>
        \textcolor{blue!70!bg}{Perfect phylogeny}
        
        \textcolor{red!70!bg}{Perfect phylogeny}
        
        $\chi_{\operatorname{PP}}(M) = 2$.
        
      \end{overprint}
    \end{columns}
  \end{example}
\end{frame}


\begin{frame}{Bad news about pp-partitions of haplotype matrices.}
  \begin{theorem}
    Finding \alert{optimal pp-partition of genotype matrices}
    is at least as hard as finding \alert{optimal colorings of
      3-uniform hypergraphs}. 
  \end{theorem}

  \begin{proof}[Proof sketch]
    \begin{enumerate}
    \item Let $G$ be a 3-uniform hypergraph.
    \item Build a matrix with a column for each vertex of $G$.
    \item For each hyperedge of $G$ add four rows inducing\\ the submatrix
      $\left(
        \begin{smallmatrix}
          2 & 2 & 2 \\
          1 & 0 & 0 \\
          0 & 1 & 0 \\
          0 & 0 & 1
        \end{smallmatrix}\right)
      $.
    \item The submatrix enforces that the three columns do not all lie
      in the same perfect phylogeny. \qedhere
    \end{enumerate}
  \end{proof}
\end{frame}

\begin{frame}{Implications for pp-partitions of genotype matrices.}
  \begin{corollary}
    Even if we know $\chi_{\operatorname{PP}}(M) = 2$ for a genotype matrix $M$,\\
    finding a pp-partition of any fixed size is still
    \begin{itemize}
    \item $\operatorname{NP}$-hard,
    \item not fixed-parameter tractable, unless
      $\operatorname{P}=\operatorname{NP}$, 
    \item very hard to approximate.
    \end{itemize}
  \end{corollary}
\end{frame}


\section{Good News: Tractability Results}

\subsection{Perfect Path Phylogenies}

\begin{frame}{Automatic optimal pp-partitioning is hopeless, but\dots}
  \begin{itemize}
  \item The hardness results are \alert{worst-case} results for\\
    \alert{highly artificial inputs}.
  \item \alert{Real biological data} might have special properties
    that make the problem \alert{tractable}.
  \item One such property is that perfect phylogenies are often
    perfect \alert{path} phylogenies:

    In HapMap data, in 70\% of the blocks where a perfect phylogeny
    is possible a perfect path phylogeny is also possible.
  \end{itemize}  
\end{frame}


\begin{frame}{Example of a perfect path phylogeny.}
  \begin{columns}[t]
    \column{.3\textwidth}
    \begin{exampleblock}{Genotype matrix}
      $G\colon$
      \begin{tabular}{ccc}
        A & B & C \\\hline
        2 & 2 & 2 \\
        0 & 2 & 0 \\
        2 & 0 & 0 \\
        0 & 2 & 2 
      \end{tabular}
    \end{exampleblock}

    \column{.3\textwidth}
    \begin{exampleblock}{Haplotype matrix}
      $H\colon$
      \begin{tabular}{ccc}
        A & B & C \\\hline
        1 & 0 & 0 \\
        0 & 1 & 1 \\
        0 & 0 & 0 \\
        0 & 1 & 0 \\
        0 & 0 & 0 \\
        1 & 0 & 0 \\
        0 & 0 & 0 \\
        0 & 1 & 1 
      \end{tabular}
    \end{exampleblock}

    \column{.4\textwidth}
    \begin{exampleblock}{Perfect path phylogeny}
      \begin{center}
        \begin{tikzpicture}[auto,thick]
          \tikzstyle{node}=%
          [%
            minimum size=10pt,%
            inner sep=0pt,%
            outer sep=0pt,%
            ball color=example text.fg,%
            circle%
          ]
        
          \node [node] {} [->]
            child {node [node] {} edge from parent node[swap]{A}}
            child {node [node] {}
              child {node [node] {} edge from parent node{C}}
              edge from parent node{B}
            };
        \end{tikzpicture}
      \end{center}
    \end{exampleblock}
  \end{columns}
\end{frame}


\begin{frame}{The modified formal computational problem.}
  We are interested in the computational complexity of \\
  the function $\chi_{\alert{\operatorname{PPP}}}$:
  \begin{itemize}
  \item It gets genotype matrices as input.
  \item It maps them to a number $k$.
  \item This number is minimal such that the sites can be
    covered by $k$ sets, each admitting a perfect \alert{path} phylogeny.
    \\
    (We call this a ppp-partition.)
  \end{itemize}
\end{frame}



\subsection{Tractability of PPP-Partitioning of Genotype Matrices}

\begin{frame}{Good news about ppp-partitions of genotype matrices.}
  \begin{theorem}
    \alert{Optimal ppp-partitions of genotype matrices} can be
    computed in \alert{polynomial time}. 
  \end{theorem}
  \begin{block}{Algorithm}
    \begin{enumerate}
    \item Build the following partial order:
      \begin{itemize}
      \item Can one column be above the other in a phylogeny?
      \item Can the columns be the two children of the root of a
        perfect path phylogeny?
      \end{itemize}
    \item Cover the partial order with as few compatible chain pairs 
      as possible. 

      For this, a maximal matching in a special graph needs to be
      computed.
    \end{enumerate}
  \end{block}
  \hyperlink{algorithm<1>}{\beamergotobutton{The algorithm in action}}
  \hypertarget{return}{}
\end{frame}

\section*{Summary}

\begin{frame}
  \frametitle<presentation>{Summary}

  \begin{itemize}
  \item
    Finding optimal pp-partitions is \alert{intractable}. 
  \item
    It is even intractable to find a pp-partition when \alert{just two 
      noncontiguous  blocks are known to suffice}.
  \item
    For perfect \alert{path} phylogenies, optimal partitions can be
    computed \alert{in polynomial time}.
  \end{itemize}
\end{frame}


\appendix

\section*{Appendix}

\begin{frame}[label=algorithm]{The algorithm in action.}{Computation of
    the partial order.}
  \begin{columns}[t]
    \column{.4\textwidth}
    \begin{exampleblock}{Genotype matrix}
      $G\colon$
      \begin{tabular}{ccccc}
        A & B & C & D & E \\\hline
        2 & 2 & 2 & 2 & 2 \\
        0 & 1 & 2 & 1 & 0 \\
        1 & 0 & 0 & 1 & 2 \\
        0 & 2 & 2 & 0 & 0
      \end{tabular}
    \end{exampleblock}
    \column{.6\textwidth}
    \begin{exampleblock}{Partial order}
      \begin{tikzpicture}[node distance=15mm]
        \tikzstyle{every node}=
        [%
          fill=green!50!black!20,%
          draw=green!50!black,%
          minimum size=7mm,%
          circle,%
          thick%
        ]

        \node (A) {A};
        \node (B) [right of=A] {B};
        \node (C) [below of=B] {C};
        \node (D) [above of=A] {D};
        \node (E) [below of=A] {E};

        \path [thick,shorten >=1pt,-stealth'] (A) edge (E)
                         (B) edge (C)
                         (D) edge (A)
                             edge[bend right] (E);

        \uncover<2>{
        \path [-,blue,thick](A) edge (B)
                                edge (C)  
                            (B) edge (E)
                            (C) edge (E);}
      \end{tikzpicture}

      Partial order: \tikz[baseline] \draw[thick,-stealth'] (0pt,.5ex)
      -- (5mm,.5ex); 

      \uncover<2>{\textcolor{blue}{Compatible as children of root:
          \tikz[baseline] \draw[thick] (0pt,.5ex) -- (5mm,.5ex);}} 
    \end{exampleblock}
  \end{columns}  
\end{frame}

\begin{frame}{The algorithm in action.}{The matching in the special graph.}
  \begin{columns}[t]
    \column{.3\textwidth}
    \begin{exampleblock}{Partial order}
      \begin{tikzpicture}[node distance=15mm]
        \tikzstyle{every node}=%
        [%
          fill=green!50!black!20,%
          draw=green!50!black,%
          minimum size=8mm,%
          circle,%
          thick%
        ]

        \node (A)              {$A$};
        \node (B) [right of=A] {$B$};
        \node (C) [below of=B] {$C$};
        \node (D) [above of=A] {$D$};
        \node (E) [below of=A] {$E$};

        \path [thick,shorten >=1pt,-stealth'] (A) edge (E)
                         (B) edge (C)
                         (D) edge (A)
                             edge[bend right] (E);

        \path [-,blue,thick](A) edge (B)
                                edge (C)  
                            (B) edge (E)
                            (C) edge (E);

        \only<3->
        {
          \path[very thick,shorten >=1pt,-stealth',red] (D) edge (A) (B) edge (C);
          \path [-,red,very thick](E) edge (B);
        }
      \end{tikzpicture}
    \end{exampleblock}
    \column{.7\textwidth}
    \begin{exampleblock}{Matching graph}
      \begin{tikzpicture}[node distance=15mm]
        \tikzstyle{every node}=%
        [%
          fill=green!50!black!20,%
          draw=green!50!black,%
          minimum size=8mm,%
          circle,%
          thick,%
          inner sep=0pt%
        ]

        \node (A)              {$A$};
        \node (B) [right of=A] {$B$};
        \node (C) [below of=B] {$C$};
        \node (D) [above of=A] {$D$};
        \node (E) [below of=A] {$E$};

        \begin{scope}[xshift=4.75cm]
          \node (A')               {$A'$};
          \node (B') [right of=A'] {$B'$};
          \node (C') [below of=B'] {$C'$};
          \node (D') [above of=A'] {$D'$};
          \node (E') [below of=A'] {$E'$};
        \end{scope}
        
        \path [thick]    (A) edge (E')
                         (B) edge (C')
                         (D) edge (A')
                             edge (E');

        \path [blue,thick](A') edge (B')
                               edge (C')  
                          (B') edge (E')
                          (C') edge (E');

        \only<2->
        {
          \path[very thick,red] (D) edge (A')
                           (B) edge (C')
                           (B') edge (E');
        }
      \end{tikzpicture}
    \end{exampleblock}
  \end{columns}

  \medskip
  \uncover<2->{A \alert{maximal matching} in the matching graph
    \uncover<3>{induces\\ \alert{perfect path phylogenies}.}}

  \hfill\hyperlink{return}{\beamerreturnbutton{Return}}
\end{frame}

\end{document}