Home Verkennen Help
Registreren Inloggen
ryan
/
resume
spiegel van https://github.com/DarwinAwardWinner/resume.git
1
0
Vork 0
Bestanden Wiki
Aftakking: main
Aftakkingen Labels
resume
/
examples
/
UVa
/
probe-selection.R.html

probe-selection.R.html 22 KB
Permalink Geschiedenis Ruwe

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206 
  1. <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN"
    2. 

  2.    "http://www.w3.org/TR/html4/strict.dtd">
    3. 

  3. 

  4. <html>
    5. 

  5. <head>
    6. 

  6.   <title></title>
    7. 

  7.   <meta http-equiv="content-type" content="text/html; charset=utf-8">
    8. 

  8.   <style type="text/css">
    9. 

  9. td.linenos { background-color: #f0f0f0; padding-right: 10px; }
    10. 

  10. span.lineno { background-color: #f0f0f0; padding: 0 5px 0 5px; }
    11. 

  11. pre { line-height: 125%; }
    12. 

  12. body .hll { background-color: #ffffcc }
    13. 

  13. body  { background: #f8f8f8; }
    14. 

  14. body .c { color: #408080; font-style: italic } /* Comment */
    15. 

  15. body .err { border: 1px solid #FF0000 } /* Error */
    16. 

  16. body .k { color: #008000; font-weight: bold } /* Keyword */
    17. 

  17. body .o { color: #666666 } /* Operator */
    18. 

  18. body .ch { color: #408080; font-style: italic } /* Comment.Hashbang */
    19. 

  19. body .cm { color: #408080; font-style: italic } /* Comment.Multiline */
    20. 

  20. body .cp { color: #BC7A00 } /* Comment.Preproc */
    21. 

  21. body .cpf { color: #408080; font-style: italic } /* Comment.PreprocFile */
    22. 

  22. body .c1 { color: #408080; font-style: italic } /* Comment.Single */
    23. 

  23. body .cs { color: #408080; font-style: italic } /* Comment.Special */
    24. 

  24. body .gd { color: #A00000 } /* Generic.Deleted */
    25. 

  25. body .ge { font-style: italic } /* Generic.Emph */
    26. 

  26. body .gr { color: #FF0000 } /* Generic.Error */
    27. 

  27. body .gh { color: #000080; font-weight: bold } /* Generic.Heading */
    28. 

  28. body .gi { color: #00A000 } /* Generic.Inserted */
    29. 

  29. body .go { color: #888888 } /* Generic.Output */
    30. 

  30. body .gp { color: #000080; font-weight: bold } /* Generic.Prompt */
    31. 

  31. body .gs { font-weight: bold } /* Generic.Strong */
    32. 

  32. body .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
    33. 

  33. body .gt { color: #0044DD } /* Generic.Traceback */
    34. 

  34. body .kc { color: #008000; font-weight: bold } /* Keyword.Constant */
    35. 

  35. body .kd { color: #008000; font-weight: bold } /* Keyword.Declaration */
    36. 

  36. body .kn { color: #008000; font-weight: bold } /* Keyword.Namespace */
    37. 

  37. body .kp { color: #008000 } /* Keyword.Pseudo */
    38. 

  38. body .kr { color: #008000; font-weight: bold } /* Keyword.Reserved */
    39. 

  39. body .kt { color: #B00040 } /* Keyword.Type */
    40. 

  40. body .m { color: #666666 } /* Literal.Number */
    41. 

  41. body .s { color: #BA2121 } /* Literal.String */
    42. 

  42. body .na { color: #7D9029 } /* Name.Attribute */
    43. 

  43. body .nb { color: #008000 } /* Name.Builtin */
    44. 

  44. body .nc { color: #0000FF; font-weight: bold } /* Name.Class */
    45. 

  45. body .no { color: #880000 } /* Name.Constant */
    46. 

  46. body .nd { color: #AA22FF } /* Name.Decorator */
    47. 

  47. body .ni { color: #999999; font-weight: bold } /* Name.Entity */
    48. 

  48. body .ne { color: #D2413A; font-weight: bold } /* Name.Exception */
    49. 

  49. body .nf { color: #0000FF } /* Name.Function */
    50. 

  50. body .nl { color: #A0A000 } /* Name.Label */
    51. 

  51. body .nn { color: #0000FF; font-weight: bold } /* Name.Namespace */
    52. 

  52. body .nt { color: #008000; font-weight: bold } /* Name.Tag */
    53. 

  53. body .nv { color: #19177C } /* Name.Variable */
    54. 

  54. body .ow { color: #AA22FF; font-weight: bold } /* Operator.Word */
    55. 

  55. body .w { color: #bbbbbb } /* Text.Whitespace */
    56. 

  56. body .mb { color: #666666 } /* Literal.Number.Bin */
    57. 

  57. body .mf { color: #666666 } /* Literal.Number.Float */
    58. 

  58. body .mh { color: #666666 } /* Literal.Number.Hex */
    59. 

  59. body .mi { color: #666666 } /* Literal.Number.Integer */
    60. 

  60. body .mo { color: #666666 } /* Literal.Number.Oct */
    61. 

  61. body .sa { color: #BA2121 } /* Literal.String.Affix */
    62. 

  62. body .sb { color: #BA2121 } /* Literal.String.Backtick */
    63. 

  63. body .sc { color: #BA2121 } /* Literal.String.Char */
    64. 

  64. body .dl { color: #BA2121 } /* Literal.String.Delimiter */
    65. 

  65. body .sd { color: #BA2121; font-style: italic } /* Literal.String.Doc */
    66. 

  66. body .s2 { color: #BA2121 } /* Literal.String.Double */
    67. 

  67. body .se { color: #BB6622; font-weight: bold } /* Literal.String.Escape */
    68. 

  68. body .sh { color: #BA2121 } /* Literal.String.Heredoc */
    69. 

  69. body .si { color: #BB6688; font-weight: bold } /* Literal.String.Interpol */
    70. 

  70. body .sx { color: #008000 } /* Literal.String.Other */
    71. 

  71. body .sr { color: #BB6688 } /* Literal.String.Regex */
    72. 

  72. body .s1 { color: #BA2121 } /* Literal.String.Single */
    73. 

  73. body .ss { color: #19177C } /* Literal.String.Symbol */
    74. 

  74. body .bp { color: #008000 } /* Name.Builtin.Pseudo */
    75. 

  75. body .fm { color: #0000FF } /* Name.Function.Magic */
    76. 

  76. body .vc { color: #19177C } /* Name.Variable.Class */
    77. 

  77. body .vg { color: #19177C } /* Name.Variable.Global */
    78. 

  78. body .vi { color: #19177C } /* Name.Variable.Instance */
    79. 

  79. body .vm { color: #19177C } /* Name.Variable.Magic */
    80. 

  80. body .il { color: #666666 } /* Literal.Number.Integer.Long */
    81. 

  81. 

  82.   </style>
    83. 

  83. </head>
    84. 

  84. <body>
    85. 

  85. <h2></h2>
    86. 

  86. 

  87. <div class="highlight"><pre><span></span><span class="c1"># We will use base 2 logarithms for now</span>
    88. 

  88. log.base <span class="o">=</span> <span class="m">2</span>
    89. 

  89. 

  90. <span class="c1"># Read the data in from the file</span>
    91. 

  91. filename.data <span class="o">=</span> <span class="s">&quot;Processed_Data_Files/Normalized_Pair_Files/All_norm_pair.txt&quot;</span>
    92. 

  92. <span class="c1">#filename.data = &quot;Raw_Data_Files/Pair_Files/All_pair.txt&quot;</span>
    93. 

  93. intensity.lin <span class="o">&lt;-</span> read.table<span class="p">(</span>file<span class="o">=</span>filename.data<span class="p">,</span> header<span class="o">=</span><span class="kc">TRUE</span><span class="p">,</span>row.names<span class="o">=</span><span class="s">&quot;PROBE_ID&quot;</span><span class="p">)</span>
    94. 

  94. 

  95. <span class="c1"># Column names and numbers used to categorize the data</span>
    96. 

  96. intensity.maincols <span class="o">=</span> <span class="kt">c</span><span class="p">(</span><span class="s">&quot;GENE_EXPR_OPTION&quot;</span><span class="p">,</span> <span class="s">&quot;SEQ_ID&quot;</span><span class="p">,</span> <span class="s">&quot;POSITION&quot;</span><span class="p">)</span>
    97. 

  97. intensity.maincolnums <span class="o">=</span> <span class="kp">which</span><span class="p">(</span><span class="kp">names</span><span class="p">(</span>intensity.lin<span class="p">)</span> <span class="o">%in%</span> intensity.maincols<span class="p">)</span>
    98. 

  98. <span class="c1"># Column names and numbers containing intensity data</span>
    99. 

  99. intensity.datacolnums <span class="o">=</span> <span class="p">(</span><span class="m">1</span><span class="o">:</span><span class="kp">dim</span><span class="p">(</span>intensity.lin<span class="p">)[</span><span class="m">2</span><span class="p">])[</span><span class="o">-</span>intensity.maincolnums<span class="p">]</span> <span class="c1"># i.e. &quot;the rest&quot;</span>
    100. 

  100. intensity.datacols <span class="o">=</span> <span class="kp">names</span><span class="p">(</span>intensity.lin<span class="p">)[</span>intensity.datacolnums<span class="p">]</span>
    101. 

  101. 

  102. <span class="c1"># Take the logarithm of the data</span>
    103. 

  103. intensity.log <span class="o">&lt;-</span> <span class="kt">data.frame</span><span class="p">(</span>intensity.lin<span class="p">[</span>intensity.maincols<span class="p">],</span><span class="kp">log</span><span class="p">(</span>intensity.lin<span class="p">[</span>intensity.datacols<span class="p">],</span> base <span class="o">=</span> log.base<span class="p">))</span>
    104. 

  104. 

  105. <span class="c1"># Separate into random (i.e. background) and data probes</span>
    106. 

  106. <span class="c1"># Split into &quot;intensity$rand&quot; and &quot;intensity$data&quot;</span>
    107. 

  107. intensity <span class="o">&lt;-</span> <span class="kp">split</span><span class="p">(</span>x <span class="o">=</span> intensity.log<span class="p">,</span> f <span class="o">=</span> <span class="p">(</span><span class="kp">ifelse</span><span class="p">(</span>intensity.log<span class="o">$</span>GENE_EXPR_OPTION <span class="o">==</span> <span class="s">&quot;RANDOM&quot;</span><span class="p">,</span><span class="s">&quot;rand&quot;</span><span class="p">,</span><span class="s">&quot;data&quot;</span><span class="p">)))</span>
    108. 

  108. <span class="kp">rm</span><span class="p">(</span>intensity.lin<span class="p">,</span> intensity.log<span class="p">)</span> <span class="c1"># Discard unused stuff to free memory</span>
    109. 

  109. 

  110. <span class="c1"># This gives a QQ plot of the data against the noise. I used it to select my cutoffs.</span>
    111. 

  111. <span class="c1">#qqplot( y=as.vector(as.matrix(intensity$data[sample(1:dim(intensity$data)[1],5000),intensity.datacols])), x=as.vector(as.matrix(intensity$rand[intensity.datacols])), ylab=&quot;Data&quot;, xlab=&quot;Random Control&quot;,main = &quot;Log10 QQ plot of Specific Probe Intensities vs. Random Controls&quot;)</span>
    112. 

  112. 

  113. <span class="c1"># Detection (low) threshold is 2 sd above mean random background</span>
    114. 

  114. intensity.rand.vector <span class="o">=</span> <span class="kp">as.vector</span><span class="p">(</span><span class="kp">as.matrix</span><span class="p">(</span>intensity<span class="o">$</span>rand<span class="p">[</span>intensity.datacols<span class="p">]))</span>
    115. 

  115. threshold.low <span class="o">=</span> <span class="kp">mean</span><span class="p">(</span>intensity.rand.vector<span class="p">)</span> <span class="o">+</span> <span class="m">2</span><span class="o">*</span>sd<span class="p">(</span>intensity.rand.vector<span class="p">)</span>
    116. 

  116. <span class="kp">rm</span><span class="p">(</span>intensity.rand.vector<span class="p">)</span>
    117. 

  117. <span class="c1"># Saturation (high) threshold is 2-fold down from max possible</span>
    118. 

  118. threshold.high <span class="o">=</span> <span class="kp">log</span><span class="p">(</span><span class="m">65535</span><span class="o">/</span><span class="m">2</span><span class="p">,</span> base<span class="o">=</span>log.base<span class="p">)</span>
    119. 

  119. <span class="kr">if</span> <span class="p">(</span>threshold.high <span class="o">&lt;=</span> threshold.low<span class="p">)</span> <span class="kp">print</span><span class="p">(</span><span class="s">&quot;Error: low threshold is too high&quot;</span><span class="p">)</span>
    120. 

  120. 

  121. <span class="c1"># Compute needed row statistics</span>
    122. 

  122. intensity<span class="o">$</span>data<span class="o">$</span>MAX <span class="o">&lt;-</span> <span class="kp">apply</span><span class="p">(</span>intensity<span class="o">$</span>data<span class="p">[</span>intensity.datacols<span class="p">],</span><span class="m">1</span><span class="p">,</span><span class="kp">max</span><span class="p">)</span>
    123. 

  123. <span class="c1"># Actually, we only need the max, it seems. Uncomment these if needed.</span>
    124. 

  124. <span class="c1">#intensity$data$MIN &lt;- apply(intensity$data[intensity.datacols],1,min)</span>
    125. 

  125. <span class="c1">#intensity$data$MEAN &lt;- rowMeans(intensity$data[intensity.datacols]) # Don&#39;t need that one</span>
    126. 

  126. <span class="c1">#intensity$data$SD &lt;- apply(intensity$data[intensity.datacols],1,sd) # Don&#39;t need that one</span>
    127. 

  127. 

  128. <span class="c1"># Sort probes into three bins: absent, present, and saturated</span>
    129. 

  129. <span class="c1"># The integer value of BIN also serves as a rank:</span>
    130. 

  130. <span class="c1"># present &lt; saturated &lt; absent; lower is better</span>
    131. 

  131. intensity<span class="o">$</span>data<span class="o">$</span>BIN <span class="o">&lt;-</span> <span class="kp">factor</span><span class="p">(</span><span class="m">1</span> <span class="o">+</span> <span class="p">(</span>intensity<span class="o">$</span>data<span class="o">$</span>MAX <span class="o">&gt;</span> threshold.high<span class="p">)</span> <span class="o">+</span> <span class="p">(</span>intensity<span class="o">$</span>data<span class="o">$</span>MAX <span class="o">&lt;</span> threshold.low<span class="p">)</span> <span class="o">*</span> <span class="m">2</span><span class="p">,</span> labels <span class="o">=</span> <span class="kt">c</span><span class="p">(</span><span class="s">&quot;present&quot;</span><span class="p">,</span> <span class="s">&quot;saturated&quot;</span><span class="p">,</span> <span class="s">&quot;absent&quot;</span><span class="p">))</span>
    132. 

  132. 

  133. <span class="c1"># Count how many probes from each probeset (SEQ_ID) went into each bin</span>
    134. 

  134. <span class="c1"># Also count total probes per set</span>
    135. 

  135. <span class="c1"># Counting is done by measuring length of data aggregated by SEQ_ID</span>
    136. 

  136. num.probes <span class="o">&lt;-</span> <span class="kp">lapply</span><span class="p">(</span><span class="kt">c</span><span class="p">(</span><span class="kt">list</span><span class="p">(</span>total<span class="o">=</span>intensity<span class="o">$</span>data<span class="p">),</span><span class="kp">split</span><span class="p">(</span>x <span class="o">=</span> intensity<span class="o">$</span>data<span class="p">,</span> f <span class="o">=</span> intensity<span class="o">$</span>data<span class="o">$</span>BIN<span class="p">)),</span> FUN <span class="o">=</span> <span class="kr">function</span> <span class="p">(</span>y<span class="p">)</span> aggregate<span class="p">(</span>x<span class="o">=</span><span class="kp">rep</span><span class="p">(</span><span class="kc">NA</span><span class="p">,</span><span class="kp">dim</span><span class="p">(</span>y<span class="p">)[</span><span class="m">1</span><span class="p">]),</span>by<span class="o">=</span><span class="kt">list</span><span class="p">(</span>SEQ_ID<span class="o">=</span>y<span class="o">$</span>SEQ_ID<span class="p">),</span>FUN<span class="o">=</span><span class="kp">length</span><span class="p">))</span>
    137. 

  137. <span class="kr">if</span><span class="p">(</span><span class="kp">mean</span><span class="p">(</span>num.probes<span class="o">$</span>present<span class="o">$</span>x<span class="p">)</span> <span class="o">&lt;</span> <span class="m">3</span><span class="p">)</span> <span class="p">{</span> <span class="kp">print</span><span class="p">(</span><span class="s">&quot;Error: Not enough present probes.&quot;</span><span class="p">)</span> <span class="p">}</span>
    138. 

  138. 

  139. 

  140. <span class="c1"># Something below this needs updating</span>
    141. 

  141. 

  142. 

  143. 

  144. <span class="c1"># 2 rankings: Nimblegen&#39;s and distance from probeset mean</span>
    145. 

  145. <span class="c1"># Nimblegen&#39;s rank is read from the design file</span>
    146. 

  146. <span class="c1"># Probeinfo file</span>
    147. 

  147. <span class="c1"># This is information parsed from the design file.</span>
    148. 

  148. <span class="c1"># The design file can&#39;t be used directly because info on probe</span>
    149. 

  149. <span class="c1"># selection is not in separate columns.</span>
    150. 

  150. filename.probeinfo <span class="o">=</span> <span class="s">&quot;Design_Files/071031_U_Va_Tobacco_Expr.probeinfo&quot;</span>
    151. 

  151. probeinfo <span class="o">&lt;-</span> read.table<span class="p">(</span>filename.probeinfo<span class="p">,</span>header<span class="o">=</span><span class="kc">TRUE</span><span class="p">,</span>row.names<span class="o">=</span><span class="s">&quot;PROBE_ID&quot;</span><span class="p">,</span>as.is<span class="o">=</span><span class="s">&quot;SEQ&quot;</span><span class="p">)</span>
    152. 

  152. <span class="c1"># Use probe names as row names for indexing</span>
    153. 

  153. <span class="c1">#row.names(probeinfo) &lt;- as.character(probeinfo$PROBE_ID)</span>
    154. 

  154. 

  155. <span class="c1"># Add Nimblegen rank to the main data frame</span>
    156. 

  156. intensity<span class="o">$</span>data<span class="p">[</span><span class="kt">c</span><span class="p">(</span><span class="s">&quot;RANK&quot;</span><span class="p">,</span><span class="s">&quot;SEQ&quot;</span><span class="p">)]</span> <span class="o">&lt;-</span> probeinfo<span class="p">[</span><span class="kp">row.names</span><span class="p">(</span>intensity<span class="o">$</span>data<span class="p">),</span><span class="kt">c</span><span class="p">(</span><span class="s">&quot;RANK&quot;</span><span class="p">,</span><span class="s">&quot;SEQ&quot;</span><span class="p">)]</span>
    157. 

  157. 

  158. 

  159. <span class="c1"># For present probes, rank is based on correlation to probeset mean</span>
    160. 

  160. 

  161. <span class="c1"># Read probeset means from calls file</span>
    162. 

  162. filename.calls <span class="o">=</span> <span class="s">&quot;Processed_Data_Files/Normalized_Calls_Files/All_norm_calls.txt&quot;</span>
    163. 

  163. calls.lin <span class="o">&lt;-</span> read.table<span class="p">(</span>filename.calls<span class="p">,</span>header<span class="o">=</span><span class="kc">TRUE</span><span class="p">,</span>row.names<span class="o">=</span><span class="s">&quot;SEQ_ID&quot;</span><span class="p">)</span>
    164. 

  164. <span class="c1"># This line simultaneously logs the data and gives the same column order as the intensity table</span>
    165. 

  165. probeset.means <span class="o">=</span> <span class="kp">log</span><span class="p">(</span>calls.lin<span class="p">[</span>intensity.datacols<span class="p">],</span> base<span class="o">=</span>log.base<span class="p">)</span>
    166. 

  166. <span class="kp">rm</span><span class="p">(</span>calls.lin<span class="p">)</span>
    167. 

  167. 

  168. <span class="c1"># Collect the relevant data into matrices for efficiency</span>
    169. 

  169. probes.present.data <span class="o">=</span> <span class="kp">t</span><span class="p">(</span>intensity<span class="o">$</span>data<span class="p">[</span>intensity<span class="o">$</span>data<span class="o">$</span>BIN <span class="o">==</span> <span class="s">&quot;present&quot;</span><span class="p">,</span>intensity.datacols<span class="p">])</span>
    170. 

  170. probeset.means.data <span class="o">=</span> <span class="kp">t</span><span class="p">(</span>probeset.means<span class="p">[</span>intensity<span class="o">$</span>data<span class="o">$</span>SEQ_ID<span class="p">[</span>intensity<span class="o">$</span>data<span class="o">$</span>BIN <span class="o">==</span> <span class="s">&quot;present&quot;</span><span class="p">],])</span>
    171. 

  171. 

  172. <span class="c1"># We invert the correlation so that lower is better</span>
    173. 

  173. intensity<span class="o">$</span>data<span class="o">$</span>RANK<span class="p">[</span>intensity<span class="o">$</span>data<span class="o">$</span>BIN <span class="o">==</span> <span class="s">&quot;present&quot;</span><span class="p">]</span> <span class="o">&lt;-</span> <span class="kp">sapply</span><span class="p">(</span><span class="m">1</span><span class="o">:</span><span class="kp">dim</span><span class="p">(</span>probes.present.data<span class="p">)[</span><span class="m">2</span><span class="p">],</span><span class="kr">function</span> <span class="p">(</span>x<span class="p">)</span> <span class="p">{</span> <span class="o">-</span>cor<span class="p">(</span>probes.present.data<span class="p">[,</span>x<span class="p">],</span>probeset.means.data<span class="p">[,</span>x<span class="p">])</span> <span class="p">})</span>
    174. 

  174. 

  175. <span class="c1"># Done with these</span>
    176. 

  176. <span class="kp">rm</span><span class="p">(</span><span class="s">&quot;probes.present.data&quot;</span><span class="p">,</span><span class="s">&quot;probeset.means.data&quot;</span><span class="p">)</span>
    177. 

  177. 

  178. <span class="c1"># Sort by bin, then rank</span>
    179. 

  179. intensity<span class="o">$</span>data.ranked <span class="o">=</span> intensity<span class="o">$</span>data<span class="p">[</span><span class="kp">order</span><span class="p">(</span>intensity<span class="o">$</span>data<span class="o">$</span>BIN<span class="p">,</span>intensity<span class="o">$</span>data<span class="o">$</span>RANK<span class="p">),]</span>
    180. 

  180. 

  181. <span class="c1"># Split by SEQ_ID</span>
    182. 

  182. probes.ranked <span class="o">&lt;-</span> <span class="kp">split</span><span class="p">(</span>x<span class="o">=</span><span class="kp">row.names</span><span class="p">(</span>intensity<span class="o">$</span>data.ranked<span class="p">),</span>
    183. 

  183.                        f<span class="o">=</span>intensity<span class="o">$</span>data.ranked<span class="o">$</span>SEQ_ID<span class="p">,</span>
    184. 

  184.                        drop<span class="o">=</span><span class="kc">TRUE</span><span class="p">)</span>
    185. 

  185. 

  186. <span class="c1"># Set the desired number of probes per sequence</span>
    187. 

  187. num.probes.desired <span class="o">&lt;-</span> <span class="m">3</span>
    188. 

  188. 

  189. <span class="c1"># A function to make any vector have length n, by truncating longer ones and padding shorter ones with NA</span>
    190. 

  190. firstN <span class="o">&lt;-</span> <span class="kr">function</span> <span class="p">(</span>v<span class="p">,</span>n<span class="p">)</span> <span class="kt">c</span><span class="p">(</span>v<span class="p">,</span><span class="kp">rep</span><span class="p">(</span><span class="kc">NA</span><span class="p">,</span>n<span class="p">))[</span><span class="m">1</span><span class="o">:</span>n<span class="p">]</span>
    191. 

  191. <span class="c1"># A function to take a string and append P1, P2, P3, etc. up to PN.</span>
    192. 

  192. probenamesN <span class="o">&lt;-</span> <span class="kr">function</span> <span class="p">(</span>s<span class="p">,</span>n<span class="p">)</span> <span class="p">(</span><span class="kp">paste</span><span class="p">(</span>s<span class="p">,</span><span class="s">&quot;P&quot;</span><span class="p">,</span><span class="m">1</span><span class="o">:</span>n<span class="p">,</span>sep<span class="o">=</span><span class="s">&quot;&quot;</span><span class="p">))[</span><span class="m">1</span><span class="o">:</span>n<span class="p">]</span>
    193. 

  193. 

  194. probes.selected <span class="o">&lt;-</span> <span class="kt">c</span><span class="p">(</span><span class="kp">sapply</span><span class="p">(</span>probes.ranked<span class="p">,</span>firstN<span class="p">,</span>num.probes.desired<span class="p">))</span>
    195. 

  195. <span class="kp">names</span><span class="p">(</span>probes.selected<span class="p">)</span> <span class="o">&lt;-</span> <span class="kt">c</span><span class="p">(</span><span class="kp">sapply</span><span class="p">(</span><span class="kp">names</span><span class="p">(</span>probes.ranked<span class="p">),</span>probenamesN<span class="p">,</span>num.probes.desired<span class="p">))</span>
    196. 

  196. <span class="c1"># Filter NA</span>
    197. 

  197. probes.selected <span class="o">&lt;-</span> probes.selected<span class="p">[</span><span class="o">!</span><span class="kp">is.na</span><span class="p">(</span>probes.selected<span class="p">)]</span>
    198. 

  198. 

  199. probes.selected.fasta <span class="o">&lt;-</span> <span class="kp">paste</span><span class="p">(</span>sep<span class="o">=</span><span class="s">&quot;&quot;</span><span class="p">,</span><span class="s">&quot;&gt;&quot;</span><span class="p">,</span><span class="kp">names</span><span class="p">(</span>probes.selected<span class="p">),</span><span class="s">&quot;\n&quot;</span><span class="p">,</span>intensity<span class="o">$</span>data<span class="p">[</span>probes.selected<span class="p">,</span><span class="s">&quot;SEQ&quot;</span><span class="p">])</span>
    200. 

  200. 

  201. <span class="kp">cat</span><span class="p">(</span>sep<span class="o">=</span><span class="s">&quot;\n&quot;</span><span class="p">,</span>probes.selected.fasta<span class="p">,</span>file<span class="o">=</span><span class="s">&quot;selected_probes.fasta&quot;</span><span class="p">)</span>
    202. 

  202. 

  203. <span class="kp">save.image</span><span class="p">(</span>file<span class="o">=</span><span class="s">&quot;probesel.rda&quot;</span><span class="p">)</span>
    204. 

  204. </pre></div>
    205. 

  205. </body>
    206. 

  206. </html>
    207.