bio-velvet_underground 0.0.1

Sign up to get free protection for your applications and to get access to all the features.
Files changed (286) hide show
  1. checksums.yaml +7 -0
  2. data/.document +5 -0
  3. data/.gitmodules +3 -0
  4. data/.travis.yml +13 -0
  5. data/Gemfile +19 -0
  6. data/LICENSE.txt +20 -0
  7. data/README.md +53 -0
  8. data/Rakefile +51 -0
  9. data/VERSION +1 -0
  10. data/ext/bioruby.patch +60 -0
  11. data/ext/mkrf_conf.rb +50 -0
  12. data/ext/src/Makefile +125 -0
  13. data/ext/src/src/allocArray.c +305 -0
  14. data/ext/src/src/allocArray.h +86 -0
  15. data/ext/src/src/autoOpen.c +107 -0
  16. data/ext/src/src/autoOpen.h +18 -0
  17. data/ext/src/src/binarySequences.c +813 -0
  18. data/ext/src/src/binarySequences.h +125 -0
  19. data/ext/src/src/concatenatedGraph.c +233 -0
  20. data/ext/src/src/concatenatedGraph.h +30 -0
  21. data/ext/src/src/concatenatedPreGraph.c +262 -0
  22. data/ext/src/src/concatenatedPreGraph.h +29 -0
  23. data/ext/src/src/correctedGraph.c +2642 -0
  24. data/ext/src/src/correctedGraph.h +32 -0
  25. data/ext/src/src/dfib.c +509 -0
  26. data/ext/src/src/dfib.h +69 -0
  27. data/ext/src/src/dfibHeap.c +89 -0
  28. data/ext/src/src/dfibHeap.h +39 -0
  29. data/ext/src/src/dfibpriv.h +105 -0
  30. data/ext/src/src/fib.c +628 -0
  31. data/ext/src/src/fib.h +78 -0
  32. data/ext/src/src/fibHeap.c +79 -0
  33. data/ext/src/src/fibHeap.h +41 -0
  34. data/ext/src/src/fibpriv.h +110 -0
  35. data/ext/src/src/globals.h +153 -0
  36. data/ext/src/src/graph.c +3983 -0
  37. data/ext/src/src/graph.h +233 -0
  38. data/ext/src/src/graphReConstruction.c +1472 -0
  39. data/ext/src/src/graphReConstruction.h +30 -0
  40. data/ext/src/src/graphStats.c +2167 -0
  41. data/ext/src/src/graphStats.h +72 -0
  42. data/ext/src/src/kmer.c +652 -0
  43. data/ext/src/src/kmer.h +73 -0
  44. data/ext/src/src/kmerOccurenceTable.c +236 -0
  45. data/ext/src/src/kmerOccurenceTable.h +44 -0
  46. data/ext/src/src/kseq.h +223 -0
  47. data/ext/src/src/locallyCorrectedGraph.c +557 -0
  48. data/ext/src/src/locallyCorrectedGraph.h +40 -0
  49. data/ext/src/src/passageMarker.c +677 -0
  50. data/ext/src/src/passageMarker.h +137 -0
  51. data/ext/src/src/preGraph.c +1717 -0
  52. data/ext/src/src/preGraph.h +106 -0
  53. data/ext/src/src/preGraphConstruction.c +990 -0
  54. data/ext/src/src/preGraphConstruction.h +26 -0
  55. data/ext/src/src/readCoherentGraph.c +557 -0
  56. data/ext/src/src/readCoherentGraph.h +30 -0
  57. data/ext/src/src/readSet.c +1734 -0
  58. data/ext/src/src/readSet.h +67 -0
  59. data/ext/src/src/recycleBin.c +199 -0
  60. data/ext/src/src/recycleBin.h +58 -0
  61. data/ext/src/src/roadMap.c +342 -0
  62. data/ext/src/src/roadMap.h +65 -0
  63. data/ext/src/src/run.c +318 -0
  64. data/ext/src/src/run.h +52 -0
  65. data/ext/src/src/run2.c +712 -0
  66. data/ext/src/src/scaffold.c +1876 -0
  67. data/ext/src/src/scaffold.h +64 -0
  68. data/ext/src/src/shortReadPairs.c +1243 -0
  69. data/ext/src/src/shortReadPairs.h +32 -0
  70. data/ext/src/src/splay.c +259 -0
  71. data/ext/src/src/splay.h +43 -0
  72. data/ext/src/src/splayTable.c +1315 -0
  73. data/ext/src/src/splayTable.h +31 -0
  74. data/ext/src/src/tightString.c +362 -0
  75. data/ext/src/src/tightString.h +82 -0
  76. data/ext/src/src/utility.c +199 -0
  77. data/ext/src/src/utility.h +98 -0
  78. data/ext/src/third-party/zlib-1.2.3/ChangeLog +855 -0
  79. data/ext/src/third-party/zlib-1.2.3/FAQ +339 -0
  80. data/ext/src/third-party/zlib-1.2.3/INDEX +51 -0
  81. data/ext/src/third-party/zlib-1.2.3/Makefile +154 -0
  82. data/ext/src/third-party/zlib-1.2.3/Makefile.in +154 -0
  83. data/ext/src/third-party/zlib-1.2.3/README +125 -0
  84. data/ext/src/third-party/zlib-1.2.3/adler32.c +149 -0
  85. data/ext/src/third-party/zlib-1.2.3/algorithm.txt +209 -0
  86. data/ext/src/third-party/zlib-1.2.3/amiga/Makefile.pup +66 -0
  87. data/ext/src/third-party/zlib-1.2.3/amiga/Makefile.sas +65 -0
  88. data/ext/src/third-party/zlib-1.2.3/as400/bndsrc +132 -0
  89. data/ext/src/third-party/zlib-1.2.3/as400/compile.clp +123 -0
  90. data/ext/src/third-party/zlib-1.2.3/as400/readme.txt +111 -0
  91. data/ext/src/third-party/zlib-1.2.3/as400/zlib.inc +331 -0
  92. data/ext/src/third-party/zlib-1.2.3/compress.c +79 -0
  93. data/ext/src/third-party/zlib-1.2.3/configure +459 -0
  94. data/ext/src/third-party/zlib-1.2.3/contrib/README.contrib +71 -0
  95. data/ext/src/third-party/zlib-1.2.3/contrib/ada/buffer_demo.adb +106 -0
  96. data/ext/src/third-party/zlib-1.2.3/contrib/ada/mtest.adb +156 -0
  97. data/ext/src/third-party/zlib-1.2.3/contrib/ada/read.adb +156 -0
  98. data/ext/src/third-party/zlib-1.2.3/contrib/ada/readme.txt +65 -0
  99. data/ext/src/third-party/zlib-1.2.3/contrib/ada/test.adb +463 -0
  100. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib-streams.adb +225 -0
  101. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib-streams.ads +114 -0
  102. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib-thin.adb +141 -0
  103. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib-thin.ads +450 -0
  104. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib.adb +701 -0
  105. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib.ads +328 -0
  106. data/ext/src/third-party/zlib-1.2.3/contrib/ada/zlib.gpr +20 -0
  107. data/ext/src/third-party/zlib-1.2.3/contrib/asm586/README.586 +43 -0
  108. data/ext/src/third-party/zlib-1.2.3/contrib/asm586/match.S +364 -0
  109. data/ext/src/third-party/zlib-1.2.3/contrib/asm686/README.686 +34 -0
  110. data/ext/src/third-party/zlib-1.2.3/contrib/asm686/match.S +329 -0
  111. data/ext/src/third-party/zlib-1.2.3/contrib/blast/Makefile +8 -0
  112. data/ext/src/third-party/zlib-1.2.3/contrib/blast/README +4 -0
  113. data/ext/src/third-party/zlib-1.2.3/contrib/blast/blast.c +444 -0
  114. data/ext/src/third-party/zlib-1.2.3/contrib/blast/blast.h +71 -0
  115. data/ext/src/third-party/zlib-1.2.3/contrib/blast/test.pk +0 -0
  116. data/ext/src/third-party/zlib-1.2.3/contrib/blast/test.txt +1 -0
  117. data/ext/src/third-party/zlib-1.2.3/contrib/delphi/ZLib.pas +557 -0
  118. data/ext/src/third-party/zlib-1.2.3/contrib/delphi/ZLibConst.pas +11 -0
  119. data/ext/src/third-party/zlib-1.2.3/contrib/delphi/readme.txt +76 -0
  120. data/ext/src/third-party/zlib-1.2.3/contrib/delphi/zlibd32.mak +93 -0
  121. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib.build +33 -0
  122. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib.chm +0 -0
  123. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib.sln +21 -0
  124. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/AssemblyInfo.cs +58 -0
  125. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/ChecksumImpl.cs +202 -0
  126. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/CircularBuffer.cs +83 -0
  127. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/CodecBase.cs +198 -0
  128. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/Deflater.cs +106 -0
  129. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/DotZLib.cs +288 -0
  130. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/DotZLib.csproj +141 -0
  131. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/GZipStream.cs +301 -0
  132. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/Inflater.cs +105 -0
  133. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/DotZLib/UnitTests.cs +274 -0
  134. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/LICENSE_1_0.txt +23 -0
  135. data/ext/src/third-party/zlib-1.2.3/contrib/dotzlib/readme.txt +58 -0
  136. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/README +1 -0
  137. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/infback9.c +608 -0
  138. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/infback9.h +37 -0
  139. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/inffix9.h +107 -0
  140. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/inflate9.h +47 -0
  141. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/inftree9.c +323 -0
  142. data/ext/src/third-party/zlib-1.2.3/contrib/infback9/inftree9.h +55 -0
  143. data/ext/src/third-party/zlib-1.2.3/contrib/inflate86/inffas86.c +1157 -0
  144. data/ext/src/third-party/zlib-1.2.3/contrib/inflate86/inffast.S +1368 -0
  145. data/ext/src/third-party/zlib-1.2.3/contrib/iostream/test.cpp +24 -0
  146. data/ext/src/third-party/zlib-1.2.3/contrib/iostream/zfstream.cpp +329 -0
  147. data/ext/src/third-party/zlib-1.2.3/contrib/iostream/zfstream.h +128 -0
  148. data/ext/src/third-party/zlib-1.2.3/contrib/iostream2/zstream.h +307 -0
  149. data/ext/src/third-party/zlib-1.2.3/contrib/iostream2/zstream_test.cpp +25 -0
  150. data/ext/src/third-party/zlib-1.2.3/contrib/iostream3/README +35 -0
  151. data/ext/src/third-party/zlib-1.2.3/contrib/iostream3/TODO +17 -0
  152. data/ext/src/third-party/zlib-1.2.3/contrib/iostream3/test.cc +50 -0
  153. data/ext/src/third-party/zlib-1.2.3/contrib/iostream3/zfstream.cc +479 -0
  154. data/ext/src/third-party/zlib-1.2.3/contrib/iostream3/zfstream.h +466 -0
  155. data/ext/src/third-party/zlib-1.2.3/contrib/masm686/match.asm +413 -0
  156. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/bld_ml64.bat +2 -0
  157. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/gvmat64.asm +513 -0
  158. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/gvmat64.obj +0 -0
  159. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/inffas8664.c +186 -0
  160. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/inffasx64.asm +392 -0
  161. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/inffasx64.obj +0 -0
  162. data/ext/src/third-party/zlib-1.2.3/contrib/masmx64/readme.txt +28 -0
  163. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/bld_ml32.bat +2 -0
  164. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/gvmat32.asm +972 -0
  165. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/gvmat32.obj +0 -0
  166. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/gvmat32c.c +62 -0
  167. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/inffas32.asm +1083 -0
  168. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/inffas32.obj +0 -0
  169. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/mkasm.bat +3 -0
  170. data/ext/src/third-party/zlib-1.2.3/contrib/masmx86/readme.txt +21 -0
  171. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/ChangeLogUnzip +67 -0
  172. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/Makefile +25 -0
  173. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/crypt.h +132 -0
  174. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/ioapi.c +177 -0
  175. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/ioapi.h +75 -0
  176. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/iowin32.c +270 -0
  177. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/iowin32.h +21 -0
  178. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/miniunz.c +585 -0
  179. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/minizip.c +420 -0
  180. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/mztools.c +281 -0
  181. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/mztools.h +31 -0
  182. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/unzip.c +1598 -0
  183. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/unzip.h +354 -0
  184. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/zip.c +1219 -0
  185. data/ext/src/third-party/zlib-1.2.3/contrib/minizip/zip.h +235 -0
  186. data/ext/src/third-party/zlib-1.2.3/contrib/pascal/example.pas +599 -0
  187. data/ext/src/third-party/zlib-1.2.3/contrib/pascal/readme.txt +76 -0
  188. data/ext/src/third-party/zlib-1.2.3/contrib/pascal/zlibd32.mak +93 -0
  189. data/ext/src/third-party/zlib-1.2.3/contrib/pascal/zlibpas.pas +236 -0
  190. data/ext/src/third-party/zlib-1.2.3/contrib/puff/Makefile +8 -0
  191. data/ext/src/third-party/zlib-1.2.3/contrib/puff/README +63 -0
  192. data/ext/src/third-party/zlib-1.2.3/contrib/puff/puff.c +837 -0
  193. data/ext/src/third-party/zlib-1.2.3/contrib/puff/puff.h +31 -0
  194. data/ext/src/third-party/zlib-1.2.3/contrib/puff/zeros.raw +0 -0
  195. data/ext/src/third-party/zlib-1.2.3/contrib/testzlib/testzlib.c +275 -0
  196. data/ext/src/third-party/zlib-1.2.3/contrib/testzlib/testzlib.txt +10 -0
  197. data/ext/src/third-party/zlib-1.2.3/contrib/untgz/Makefile +14 -0
  198. data/ext/src/third-party/zlib-1.2.3/contrib/untgz/Makefile.msc +17 -0
  199. data/ext/src/third-party/zlib-1.2.3/contrib/untgz/untgz.c +674 -0
  200. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/readme.txt +73 -0
  201. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/miniunz.vcproj +126 -0
  202. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/minizip.vcproj +126 -0
  203. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/testzlib.vcproj +126 -0
  204. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/zlib.rc +32 -0
  205. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/zlibstat.vcproj +246 -0
  206. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/zlibvc.def +92 -0
  207. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/zlibvc.sln +78 -0
  208. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc7/zlibvc.vcproj +445 -0
  209. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/miniunz.vcproj +566 -0
  210. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/minizip.vcproj +563 -0
  211. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/testzlib.vcproj +948 -0
  212. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/testzlibdll.vcproj +567 -0
  213. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/zlib.rc +32 -0
  214. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/zlibstat.vcproj +870 -0
  215. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/zlibvc.def +92 -0
  216. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/zlibvc.sln +144 -0
  217. data/ext/src/third-party/zlib-1.2.3/contrib/vstudio/vc8/zlibvc.vcproj +1219 -0
  218. data/ext/src/third-party/zlib-1.2.3/crc32.c +423 -0
  219. data/ext/src/third-party/zlib-1.2.3/crc32.h +441 -0
  220. data/ext/src/third-party/zlib-1.2.3/deflate.c +1736 -0
  221. data/ext/src/third-party/zlib-1.2.3/deflate.h +331 -0
  222. data/ext/src/third-party/zlib-1.2.3/example.c +565 -0
  223. data/ext/src/third-party/zlib-1.2.3/examples/README.examples +42 -0
  224. data/ext/src/third-party/zlib-1.2.3/examples/fitblk.c +233 -0
  225. data/ext/src/third-party/zlib-1.2.3/examples/gun.c +693 -0
  226. data/ext/src/third-party/zlib-1.2.3/examples/gzappend.c +500 -0
  227. data/ext/src/third-party/zlib-1.2.3/examples/gzjoin.c +448 -0
  228. data/ext/src/third-party/zlib-1.2.3/examples/gzlog.c +413 -0
  229. data/ext/src/third-party/zlib-1.2.3/examples/gzlog.h +58 -0
  230. data/ext/src/third-party/zlib-1.2.3/examples/zlib_how.html +523 -0
  231. data/ext/src/third-party/zlib-1.2.3/examples/zpipe.c +191 -0
  232. data/ext/src/third-party/zlib-1.2.3/examples/zran.c +404 -0
  233. data/ext/src/third-party/zlib-1.2.3/gzio.c +1026 -0
  234. data/ext/src/third-party/zlib-1.2.3/infback.c +623 -0
  235. data/ext/src/third-party/zlib-1.2.3/inffast.c +318 -0
  236. data/ext/src/third-party/zlib-1.2.3/inffast.h +11 -0
  237. data/ext/src/third-party/zlib-1.2.3/inffixed.h +94 -0
  238. data/ext/src/third-party/zlib-1.2.3/inflate.c +1368 -0
  239. data/ext/src/third-party/zlib-1.2.3/inflate.h +115 -0
  240. data/ext/src/third-party/zlib-1.2.3/inftrees.c +329 -0
  241. data/ext/src/third-party/zlib-1.2.3/inftrees.h +55 -0
  242. data/ext/src/third-party/zlib-1.2.3/make_vms.com +461 -0
  243. data/ext/src/third-party/zlib-1.2.3/minigzip.c +322 -0
  244. data/ext/src/third-party/zlib-1.2.3/msdos/Makefile.bor +109 -0
  245. data/ext/src/third-party/zlib-1.2.3/msdos/Makefile.dj2 +104 -0
  246. data/ext/src/third-party/zlib-1.2.3/msdos/Makefile.emx +69 -0
  247. data/ext/src/third-party/zlib-1.2.3/msdos/Makefile.msc +106 -0
  248. data/ext/src/third-party/zlib-1.2.3/msdos/Makefile.tc +94 -0
  249. data/ext/src/third-party/zlib-1.2.3/old/Makefile.riscos +151 -0
  250. data/ext/src/third-party/zlib-1.2.3/old/README +3 -0
  251. data/ext/src/third-party/zlib-1.2.3/old/descrip.mms +48 -0
  252. data/ext/src/third-party/zlib-1.2.3/old/os2/Makefile.os2 +136 -0
  253. data/ext/src/third-party/zlib-1.2.3/old/os2/zlib.def +51 -0
  254. data/ext/src/third-party/zlib-1.2.3/old/visual-basic.txt +160 -0
  255. data/ext/src/third-party/zlib-1.2.3/old/zlib.html +971 -0
  256. data/ext/src/third-party/zlib-1.2.3/projects/README.projects +41 -0
  257. data/ext/src/third-party/zlib-1.2.3/projects/visualc6/README.txt +73 -0
  258. data/ext/src/third-party/zlib-1.2.3/projects/visualc6/example.dsp +278 -0
  259. data/ext/src/third-party/zlib-1.2.3/projects/visualc6/minigzip.dsp +278 -0
  260. data/ext/src/third-party/zlib-1.2.3/projects/visualc6/zlib.dsp +609 -0
  261. data/ext/src/third-party/zlib-1.2.3/projects/visualc6/zlib.dsw +59 -0
  262. data/ext/src/third-party/zlib-1.2.3/qnx/package.qpg +141 -0
  263. data/ext/src/third-party/zlib-1.2.3/trees.c +1219 -0
  264. data/ext/src/third-party/zlib-1.2.3/trees.h +128 -0
  265. data/ext/src/third-party/zlib-1.2.3/uncompr.c +61 -0
  266. data/ext/src/third-party/zlib-1.2.3/win32/DLL_FAQ.txt +397 -0
  267. data/ext/src/third-party/zlib-1.2.3/win32/Makefile.bor +107 -0
  268. data/ext/src/third-party/zlib-1.2.3/win32/Makefile.emx +69 -0
  269. data/ext/src/third-party/zlib-1.2.3/win32/Makefile.gcc +141 -0
  270. data/ext/src/third-party/zlib-1.2.3/win32/Makefile.msc +126 -0
  271. data/ext/src/third-party/zlib-1.2.3/win32/VisualC.txt +3 -0
  272. data/ext/src/third-party/zlib-1.2.3/win32/zlib.def +60 -0
  273. data/ext/src/third-party/zlib-1.2.3/win32/zlib1.rc +39 -0
  274. data/ext/src/third-party/zlib-1.2.3/zconf.h +332 -0
  275. data/ext/src/third-party/zlib-1.2.3/zconf.in.h +332 -0
  276. data/ext/src/third-party/zlib-1.2.3/zlib.3 +159 -0
  277. data/ext/src/third-party/zlib-1.2.3/zlib.h +1357 -0
  278. data/ext/src/third-party/zlib-1.2.3/zutil.c +318 -0
  279. data/ext/src/third-party/zlib-1.2.3/zutil.h +269 -0
  280. data/lib/bio-velvet_underground.rb +12 -0
  281. data/lib/bio-velvet_underground/external/VERSION +1 -0
  282. data/lib/bio-velvet_underground/velvet_underground.rb +72 -0
  283. data/spec/binary_sequence_store_spec.rb +27 -0
  284. data/spec/data/1/CnyUnifiedSeq +0 -0
  285. data/spec/spec_helper.rb +31 -0
  286. metadata +456 -0
@@ -0,0 +1,1876 @@
1
+ /*
2
+ Copyright 2007, 2008 Daniel Zerbino (zerbino@ebi.ac.uk)
3
+
4
+ This file is part of Velvet.
5
+
6
+ Velvet is free software; you can redistribute it and/or modify
7
+ it under the terms of the GNU General Public License as published by
8
+ the Free Software Foundation; either version 2 of the License, or
9
+ (at your option) any later version.
10
+
11
+ Velvet is distributed in the hope that it will be useful,
12
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
13
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14
+ GNU General Public License for more details.
15
+
16
+ You should have received a copy of the GNU General Public License
17
+ along with Velvet; if not, write to the Free Software
18
+ Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
19
+
20
+ */
21
+ #include <stdlib.h>
22
+ #include <stdio.h>
23
+ #include <time.h>
24
+ #include <math.h>
25
+ #include <sys/time.h>
26
+
27
+ #ifdef _OPENMP
28
+ #include <omp.h>
29
+ #endif
30
+
31
+ #include "globals.h"
32
+ #include "graph.h"
33
+ #include "concatenatedGraph.h"
34
+ #include "recycleBin.h"
35
+ #include "locallyCorrectedGraph.h"
36
+ #include "passageMarker.h"
37
+ #include "readSet.h"
38
+ #include "utility.h"
39
+ #include "scaffold.h"
40
+
41
+ #define BLOCK_SIZE 100000
42
+ #define LN2 1.4
43
+
44
+ static int PEBBLE_ROUND_NUM = 0;
45
+
46
+ typedef struct readOccurence_st ReadOccurence;
47
+ static double paired_exp_fraction = 0.1;
48
+
49
+ struct connection_st {
50
+ Node *destination;
51
+ Connection *right;
52
+ Connection *left;
53
+ Connection *twin;
54
+ float distance;
55
+ float variance;
56
+ IDnum direct_count;
57
+ IDnum paired_count;
58
+ unsigned char clean;
59
+ } ATTRIBUTE_PACKED;
60
+
61
+ struct readOccurence_st {
62
+ IDnum position;
63
+ IDnum offset;
64
+ IDnum nodeID;
65
+ } ATTRIBUTE_PACKED;
66
+
67
+ // Global params
68
+ static IDnum UNRELIABLE_CONNECTION_CUTOFF = 5;
69
+
70
+ // Global pointers
71
+ static Graph *graph;
72
+ static Connection **scaffold = NULL;
73
+ static RecycleBin *connectionMemory = NULL;
74
+ static boolean estimated[CATEGORIES + 1];
75
+
76
+ #ifdef _OPENMP
77
+
78
+ #define READS_PER_LOCK 32
79
+
80
+
81
+ /* Array of reads locks */
82
+ static omp_lock_t *readsLocks = NULL;
83
+ /* Array of per-node locks */
84
+ static omp_lock_t *nodeLocks = NULL;
85
+
86
+ static void
87
+ createReadsLocks()
88
+ {
89
+ Coordinate nbLocks;
90
+ Coordinate lockIndex;
91
+
92
+ if (readsLocks)
93
+ free (readsLocks);
94
+ nbLocks = 1 + sequenceCount(graph) / READS_PER_LOCK;
95
+ readsLocks = mallocOrExit(nbLocks, omp_lock_t);
96
+
97
+ #pragma omp parallel for
98
+ for (lockIndex = 0; lockIndex < nbLocks; lockIndex++)
99
+ omp_init_lock(readsLocks + lockIndex);
100
+ }
101
+
102
+ static inline void lockRead(IDnum readID)
103
+ {
104
+ omp_set_lock (readsLocks + readID / READS_PER_LOCK);
105
+ }
106
+
107
+ static inline void unLockRead(IDnum readID)
108
+ {
109
+ omp_unset_lock (readsLocks + readID / READS_PER_LOCK);
110
+ }
111
+
112
+ static void
113
+ createNodeLocks(Graph *graph)
114
+ {
115
+ IDnum nbNodes;
116
+ IDnum nodeIndex;
117
+
118
+ nbNodes = nodeCount(graph) + 1;
119
+ if (nodeLocks)
120
+ free (nodeLocks);
121
+ nodeLocks = mallocOrExit(nbNodes, omp_lock_t);
122
+
123
+ #pragma omp parallel for
124
+ for (nodeIndex = 0; nodeIndex < nbNodes; nodeIndex++)
125
+ omp_init_lock(nodeLocks + nodeIndex);
126
+ }
127
+
128
+ /* Tries to avoid deadlocking */
129
+ static inline void lockTwoNodes(IDnum nodeID, IDnum node2ID)
130
+ {
131
+ if (nodeID < 0)
132
+ nodeID = -nodeID;
133
+ if (node2ID < 0)
134
+ node2ID = -node2ID;
135
+
136
+ /* Lock lowest ID first to avoid deadlocks */
137
+ if (nodeID < node2ID)
138
+ {
139
+ omp_set_lock (nodeLocks + nodeID);
140
+ omp_set_lock (nodeLocks + node2ID);
141
+ }
142
+ else
143
+ {
144
+ omp_set_lock (nodeLocks + node2ID);
145
+ omp_set_lock (nodeLocks + nodeID);
146
+ }
147
+ }
148
+
149
+ static inline void unLockTwoNodes(IDnum nodeID, IDnum node2ID)
150
+ {
151
+ if (nodeID < 0)
152
+ nodeID = -nodeID;
153
+ if (node2ID < 0)
154
+ node2ID = -node2ID;
155
+
156
+ omp_unset_lock (nodeLocks + nodeID);
157
+ omp_unset_lock (nodeLocks + node2ID);
158
+ }
159
+ #endif
160
+
161
+ static Connection *allocateConnection()
162
+ {
163
+ Connection *connect;
164
+ #ifdef _OPENMP
165
+ #pragma omp critical
166
+ {
167
+ #endif
168
+ if (connectionMemory == NULL)
169
+ connectionMemory =
170
+ newRecycleBin(sizeof(Connection), BLOCK_SIZE);
171
+
172
+ connect = allocatePointer(connectionMemory);
173
+ #ifdef _OPENMP
174
+ }
175
+ #endif
176
+ connect->destination = NULL;
177
+ connect->clean = false;
178
+ return connect;
179
+ }
180
+
181
+ static void deallocateConnection(Connection * connect)
182
+ {
183
+ deallocatePointer(connectionMemory, connect);
184
+ }
185
+
186
+ Node * getConnectionDestination(Connection * connect) {
187
+ return connect->destination;
188
+ }
189
+
190
+ Connection * getNextConnection(Connection * connect) {
191
+ return connect->right;
192
+ }
193
+
194
+ Connection * getTwinConnection(Connection * connect) {
195
+ return connect->twin;
196
+ }
197
+
198
+ Coordinate getConnectionDistance(Connection * connect) {
199
+ return (Coordinate) connect->distance;
200
+ }
201
+
202
+ double getConnectionVariance(Connection * connect) {
203
+ return connect->variance;
204
+ }
205
+
206
+ IDnum getConnectionDirectCount(Connection * connect) {
207
+ return connect->direct_count;
208
+ }
209
+
210
+ IDnum getConnectionPairedCount(Connection * connect) {
211
+ return connect->paired_count;
212
+ }
213
+
214
+ Connection * getConnection(Node * node) {
215
+ return scaffold[getNodeID(node) + nodeCount(graph)];
216
+ }
217
+
218
+ void incrementConnectionDistance(Connection * connect, Coordinate increment) {
219
+ connect->distance += increment;
220
+ }
221
+
222
+ static double norm(double X)
223
+ {
224
+ return 0.4 * exp(-X * X / 2);
225
+ }
226
+
227
+ static double normInt(double X, double Y)
228
+ {
229
+ return (erf(0.7 * Y) - erf(0.7 * X)) / 2;
230
+ }
231
+
232
+ static IDnum expectedNumberOfConnections(IDnum IDA, Connection * connect,
233
+ IDnum ** counts, Category cat)
234
+ {
235
+ Node *A = getNodeInGraph(graph, IDA);
236
+ Node *B = connect->destination;
237
+ double left, middle, right;
238
+ Coordinate longLength, shortLength, D;
239
+ IDnum longCount;
240
+ double M, N, O, P;
241
+ Coordinate mu = getInsertLength(graph, cat);
242
+ double sigma = sqrt(getInsertLength_var(graph, cat));
243
+ double result;
244
+
245
+ if (mu <= 0)
246
+ return 0;
247
+
248
+ if (getNodeLength(A) < getNodeLength(B)) {
249
+ longLength = getNodeLength(B);
250
+ shortLength = getNodeLength(A);
251
+ longCount = counts[cat][getNodeID(B) + nodeCount(graph)];
252
+ } else {
253
+ longLength = getNodeLength(A);
254
+ shortLength = getNodeLength(B);
255
+ longCount = counts[cat][IDA + nodeCount(graph)];
256
+ }
257
+
258
+ D = getConnectionDistance(connect) - (longLength + shortLength) / 2;
259
+
260
+ M = (D - mu) / sigma;
261
+ N = (D + shortLength - mu) / sigma;
262
+ O = (D + longLength - mu) / sigma;
263
+ P = (D + shortLength + longLength - mu) / sigma;
264
+
265
+ left = ((norm(M) - norm(N)) - M * normInt(M, N)) * sigma;
266
+ middle = shortLength * normInt(N, O);
267
+ right = ((norm(O) - norm(P)) - P * normInt(O, P)) * (-sigma);
268
+
269
+ result = (longCount * (left + middle + right)) / longLength;
270
+
271
+ if (result > 0)
272
+ return (IDnum) result;
273
+ else
274
+ return 0;
275
+ }
276
+
277
+ void destroyConnection(Connection * connect, IDnum nodeID)
278
+ {
279
+ Connection *previous, *next;
280
+
281
+ //velvetLog("Destroying connection from %li to %li\n", nodeID, getNodeID(connect->destination));
282
+
283
+ if (connect == NULL)
284
+ return;
285
+
286
+ previous = connect->left;
287
+ next = connect->right;
288
+
289
+ if (previous != NULL)
290
+ previous->right = next;
291
+ if (next != NULL)
292
+ next->left = previous;
293
+
294
+ if (scaffold[nodeID + nodeCount(graph)] == connect)
295
+ scaffold[nodeID + nodeCount(graph)] = next;
296
+
297
+ if (connect->twin != NULL) {
298
+ connect->twin->twin = NULL;
299
+ destroyConnection(connect->twin,
300
+ getNodeID(connect->destination));
301
+ }
302
+
303
+ deallocateConnection(connect);
304
+ }
305
+
306
+ static boolean testConnection(IDnum IDA,
307
+ Connection *connect,
308
+ IDnum **counts,
309
+ boolean *shadows)
310
+ {
311
+ IDnum total = 0;
312
+ Category cat;
313
+
314
+ // Spare unique -> undetermined node connections
315
+ if (!getUniqueness(connect->destination))
316
+ return true;
317
+
318
+ // Destroy tenuous connections
319
+ if (connect->paired_count + connect->direct_count <
320
+ UNRELIABLE_CONNECTION_CUTOFF)
321
+ return false;
322
+
323
+ for (cat = 0; cat < CATEGORIES; cat++)
324
+ if (!shadows[cat] || cat <= PEBBLE_ROUND_NUM)
325
+ total += expectedNumberOfConnections(IDA, connect, counts, cat);
326
+
327
+ // Remove inconsistent connections
328
+ return connect->paired_count >= total * paired_exp_fraction;
329
+ }
330
+
331
+ static IDnum *computeReadToNodeCounts(Coordinate *totalCount)
332
+ {
333
+ IDnum nodeIndex;
334
+ IDnum maxNodeIndex = 2 * nodeCount(graph) + 1;
335
+ IDnum maxReadIndex = sequenceCount(graph) + 1;
336
+ IDnum *readNodeCounts = callocOrExit(maxReadIndex, IDnum);
337
+ unsigned char *readMarker = callocOrExit(1 + maxReadIndex / 8, unsigned char);
338
+ Coordinate total = 0;
339
+
340
+ velvetLog("Computing read to node mapping array sizes\n");
341
+
342
+ #ifdef _OPENMP
343
+ #pragma omp parallel for reduction(+:total)
344
+ #endif
345
+ for (nodeIndex = 0; nodeIndex < maxNodeIndex; nodeIndex++) {
346
+ Node *node;
347
+ ShortReadMarker *nodeArray;
348
+ IDnum nodeReadCount;
349
+ IDnum readIndex;
350
+
351
+ node = getNodeInGraph(graph, nodeIndex - nodeCount(graph));
352
+ if (node == NULL)
353
+ continue;
354
+ nodeArray = getNodeReads(node, graph);
355
+ nodeReadCount = getNodeReadCount(node, graph);
356
+
357
+ // Short reads
358
+ for (readIndex = 0; readIndex < nodeReadCount; readIndex++) {
359
+ ShortReadMarker *shortMarker;
360
+ IDnum readID;
361
+
362
+ shortMarker = getShortReadMarkerAtIndex(nodeArray,
363
+ readIndex);
364
+ readID = getShortReadMarkerID(shortMarker);
365
+ #ifdef _OPENMP
366
+ #pragma omp atomic
367
+ #endif
368
+ readNodeCounts[readID]++;
369
+ total++;
370
+ }
371
+ }
372
+
373
+ for (nodeIndex = 0; nodeIndex < maxNodeIndex; nodeIndex++) {
374
+ Node *node;
375
+ PassageMarkerI marker;
376
+
377
+ node = getNodeInGraph(graph, nodeIndex - nodeCount(graph));
378
+ if (node == NULL)
379
+ continue;
380
+ // Long reads
381
+ for (marker = getMarker(node); marker != NULL_IDX;
382
+ marker = getNextInNode(marker)) {
383
+ IDnum readIndex = getPassageMarkerSequenceID(marker);;
384
+
385
+ if (readIndex < 0)
386
+ continue;
387
+
388
+ const unsigned int idx = readIndex / 8;
389
+ const unsigned int mask = 1 << (readIndex & 7);
390
+ if (readMarker[idx] & mask)
391
+ continue;
392
+
393
+ readNodeCounts[readIndex]++;
394
+ total++;
395
+ readMarker[idx] |= mask;
396
+ }
397
+
398
+ // Clean up marker array
399
+ for (marker = getMarker(node); marker != NULL_IDX;
400
+ marker = getNextInNode(marker)) {
401
+ IDnum readIndex = getPassageMarkerSequenceID(marker);
402
+ if (readIndex > 0)
403
+ // No need to go bit-wise
404
+ readMarker[readIndex / 8] = 0;
405
+ }
406
+ }
407
+
408
+ *totalCount = total;
409
+ free(readMarker);
410
+ return readNodeCounts;
411
+ }
412
+
413
+ static ReadOccurence **allocateReadToNodeTables(IDnum * readNodeCounts,
414
+ Coordinate totalCount,
415
+ ReadOccurence **readNodesArray)
416
+ {
417
+ Coordinate offset = 0;
418
+ IDnum readIndex;
419
+ IDnum maxReadIndex = sequenceCount(graph) + 1;
420
+ ReadOccurence **readNodes = callocOrExit(maxReadIndex, ReadOccurence *);
421
+ *readNodesArray = callocOrExit(totalCount, ReadOccurence);
422
+
423
+ for (readIndex = 1; readIndex < maxReadIndex; readIndex++) {
424
+ if (readNodeCounts[readIndex] != 0) {
425
+ readNodes[readIndex] = *readNodesArray + offset;
426
+ offset += readNodeCounts[readIndex];
427
+ readNodeCounts[readIndex] = 0;
428
+ }
429
+ }
430
+
431
+ return readNodes;
432
+ }
433
+
434
+ static void computePartialReadToNodeMappingShort(IDnum nodeID,
435
+ ReadOccurence ** readNodes,
436
+ IDnum * readNodeCounts)
437
+ {
438
+ ShortReadMarker *shortMarker;
439
+ IDnum index, readIndex;
440
+ ReadOccurence *readArray, *readOccurence;
441
+ Node *node = getNodeInGraph(graph, nodeID);
442
+ ShortReadMarker *nodeArray = getNodeReads(node, graph);
443
+ IDnum nodeReadCount = getNodeReadCount(node, graph);
444
+
445
+ for (index = 0; index < nodeReadCount; index++) {
446
+ shortMarker = getShortReadMarkerAtIndex(nodeArray, index);
447
+ readIndex = getShortReadMarkerID(shortMarker);
448
+ readArray = readNodes[readIndex];
449
+ #ifdef _OPENMP
450
+ lockRead(readIndex);
451
+ #endif
452
+ readOccurence = &readArray[readNodeCounts[readIndex]];
453
+ readOccurence->nodeID = nodeID;
454
+ readOccurence->position =
455
+ getShortReadMarkerPosition(shortMarker);
456
+ readOccurence->offset =
457
+ getShortReadMarkerOffset(shortMarker);
458
+ readNodeCounts[readIndex]++;
459
+ #ifdef _OPENMP
460
+ unLockRead(readIndex);
461
+ #endif
462
+ }
463
+ }
464
+
465
+ static void computePartialReadToNodeMappingLong(IDnum nodeID,
466
+ ReadOccurence ** readNodes,
467
+ IDnum * readNodeCounts,
468
+ unsigned char *readMarker,
469
+ ReadSet * reads)
470
+ {
471
+ IDnum readIndex;
472
+ ReadOccurence *readArray, *readOccurence;
473
+ Node *node = getNodeInGraph(graph, nodeID);
474
+ PassageMarkerI marker;
475
+
476
+ for (marker = getMarker(node); marker != NULL_IDX;
477
+ marker = getNextInNode(marker)) {
478
+ readIndex = getPassageMarkerSequenceID(marker);
479
+ if (readIndex <= 0 || reads->categories[readIndex - 1] == REFERENCE)
480
+ continue;
481
+
482
+ const unsigned int idx = readIndex / 8;
483
+ const unsigned int mask = 1 << (readIndex & 7);
484
+ if (readMarker[idx] & mask) {
485
+ readArray = readNodes[readIndex];
486
+ readOccurence =
487
+ &readArray[readNodeCounts[readIndex] - 1];
488
+ readOccurence->position = -1;
489
+ readOccurence->offset = -1;
490
+ } else {
491
+ readArray = readNodes[readIndex];
492
+ readOccurence =
493
+ &readArray[readNodeCounts[readIndex]];
494
+ readOccurence->nodeID = nodeID;
495
+ readOccurence->position = getStartOffset(marker);
496
+ readOccurence->offset =
497
+ getPassageMarkerStart(marker);
498
+ readNodeCounts[readIndex]++;
499
+ readMarker[idx] |= mask;
500
+ }
501
+ }
502
+
503
+ for (marker = getMarker(node); marker != NULL_IDX;
504
+ marker = getNextInNode(marker)) {
505
+ readIndex = getPassageMarkerSequenceID(marker);
506
+ if (readIndex > 0)
507
+ // No need to go bit-wise
508
+ readMarker[readIndex / 8] = 0;
509
+ }
510
+ }
511
+
512
+ static ReadOccurence **computeReadToNodeMappings(IDnum * readNodeCounts,
513
+ ReadSet * reads,
514
+ Coordinate totalCount,
515
+ ReadOccurence **readNodesArray)
516
+ {
517
+ unsigned char *readMarker;
518
+ IDnum nodeID;
519
+ IDnum nodes = nodeCount(graph);
520
+ ReadOccurence **readNodes = allocateReadToNodeTables(readNodeCounts,
521
+ totalCount,
522
+ readNodesArray);
523
+
524
+ velvetLog("Computing read to node mappings\n");
525
+
526
+ #ifdef _OPENMP
527
+ createReadsLocks();
528
+ #pragma omp parallel for
529
+ #endif
530
+ for (nodeID = -nodes; nodeID <= nodes; nodeID++)
531
+ if (nodeID != 0 && getNodeInGraph(graph, nodeID))
532
+ computePartialReadToNodeMappingShort(nodeID, readNodes,
533
+ readNodeCounts);
534
+
535
+ #ifdef _OPENMP
536
+ free(readsLocks);
537
+ readsLocks = NULL;
538
+ #endif
539
+
540
+ readMarker = callocOrExit(1 + sequenceCount(graph) / 8, unsigned char);
541
+ for (nodeID = -nodes; nodeID <= nodes; nodeID++)
542
+ if (nodeID != 0 && getNodeInGraph(graph, nodeID))
543
+ computePartialReadToNodeMappingLong(nodeID, readNodes,
544
+ readNodeCounts,
545
+ readMarker,
546
+ reads);
547
+
548
+ free(readMarker);
549
+ return readNodes;
550
+ }
551
+
552
+ static unsigned char * countCoOccurences(IDnum * coOccurencesCount,
553
+ ReadOccurence ** readNodes,
554
+ IDnum * readNodeCounts,
555
+ IDnum * readPairs,
556
+ Category * cats)
557
+ {
558
+ IDnum readIndex, readPairIndex;
559
+ IDnum readNodeCount;
560
+ IDnum readOccurenceIndex, readPairOccurenceIndex;
561
+ ReadOccurence * readOccurence, *readPairOccurence;
562
+ unsigned char *interestingReads = callocOrExit(1 + sequenceCount(graph) / 8, unsigned char);
563
+ Category libID;
564
+
565
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
566
+ coOccurencesCount[libID] = 0;
567
+
568
+ for (readIndex = 0; readIndex < sequenceCount(graph); readIndex++) {
569
+ // Eliminating dodgy, unpaired, already counted or user-specified reads
570
+ if ( readPairs[readIndex] < readIndex
571
+ || getInsertLength(graph, cats[readIndex]) > -1)
572
+ continue;
573
+
574
+ // Check for co-occurence
575
+ // We know that for each read the read occurences are ordered by increasing node ID
576
+ // Therefore one list is followed by increasing index, whereas the other is followed
577
+ // by decreasing index
578
+ libID = cats[readIndex] / 2;
579
+ readPairIndex = readPairs[readIndex];
580
+
581
+ readOccurenceIndex = 0;
582
+ readOccurence = readNodes[readIndex + 1];
583
+ readNodeCount = readNodeCounts[readIndex + 1];
584
+
585
+ readPairOccurenceIndex = readNodeCounts[readPairIndex + 1] - 1;
586
+ readPairOccurence = &(readNodes[readPairIndex + 1][readPairOccurenceIndex]);
587
+
588
+ while (readOccurenceIndex < readNodeCount && readPairOccurenceIndex >= 0) {
589
+ if (readOccurence->nodeID == -readPairOccurence->nodeID) {
590
+ if (readOccurence->position > 0 && readPairOccurence->position > 0) {
591
+ coOccurencesCount[libID]++;
592
+ interestingReads[readIndex / 8] |= 1 << (readIndex & 7);
593
+ break;
594
+ } else {
595
+ readOccurence++;
596
+ readOccurenceIndex++;
597
+ readPairOccurence--;
598
+ readPairOccurenceIndex--;
599
+ }
600
+ } else if (readOccurence->nodeID < -readPairOccurence->nodeID) {
601
+ readOccurence++;
602
+ readOccurenceIndex++;
603
+ } else {
604
+ readPairOccurence--;
605
+ readPairOccurenceIndex--;
606
+ }
607
+ }
608
+ }
609
+
610
+ return interestingReads;
611
+ }
612
+
613
+ static void measureCoOccurences(IDnum ** coOccurences,
614
+ unsigned char * interestingReads,
615
+ ReadOccurence ** readNodes,
616
+ IDnum * readNodeCounts,
617
+ IDnum * readPairs,
618
+ Category * cats)
619
+ {
620
+ IDnum coOccurencesIndex[CATEGORIES + 1];
621
+ IDnum observationIndex;
622
+ IDnum readIndex, readPairIndex;
623
+ IDnum readNodeCount;
624
+ IDnum readOccurenceIndex, readPairOccurenceIndex;
625
+ ReadOccurence * readOccurence, *readPairOccurence;
626
+ Category libID;
627
+
628
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
629
+ coOccurencesIndex[libID] = 0;
630
+
631
+ for (readIndex = 0; readIndex < sequenceCount(graph); readIndex++) {
632
+ // Eliminating dodgy, unpaired, already counted or user-specified reads
633
+ if (!(interestingReads[readIndex / 8] & (1 << (readIndex & 7))))
634
+ continue;
635
+
636
+ // Find co-occurence
637
+ // We know that for each read the read occurences are ordered by increasing node ID
638
+ libID = cats[readIndex]/2;
639
+ readPairIndex = readPairs[readIndex];
640
+ observationIndex = coOccurencesIndex[libID];
641
+
642
+ readOccurence = readNodes[readIndex + 1];
643
+ readOccurenceIndex = 0;
644
+ readNodeCount = readNodeCounts[readIndex + 1];
645
+
646
+ readPairOccurenceIndex = readNodeCounts[readPairIndex + 1] - 1;
647
+ readPairOccurence = &(readNodes[readPairIndex + 1][readPairOccurenceIndex]);
648
+
649
+ while (readOccurenceIndex < readNodeCount && readPairOccurenceIndex >= 0) {
650
+ if (readOccurence->nodeID == -readPairOccurence->nodeID) {
651
+ if (readOccurence->position > 0 && readPairOccurence->position > 0) {
652
+ coOccurences[libID][observationIndex] =
653
+ getNodeLength(getNodeInGraph(graph, readOccurence->nodeID))
654
+ + getWordLength(graph) - 1
655
+ - (readOccurence->position - readOccurence->offset)
656
+ - (readPairOccurence->position - readPairOccurence->offset);
657
+ coOccurencesIndex[libID]++;
658
+ break;
659
+ } else {
660
+ readOccurence++;
661
+ readOccurenceIndex++;
662
+ readPairOccurence--;
663
+ readPairOccurenceIndex--;
664
+ }
665
+ } else if (readOccurence->nodeID < -readPairOccurence->nodeID) {
666
+ readOccurence++;
667
+ readOccurenceIndex++;
668
+ } else {
669
+ readPairOccurence--;
670
+ readPairOccurenceIndex--;
671
+ }
672
+ }
673
+ }
674
+ }
675
+
676
+ int compareReadOccurences(const void *A, const void * B) {
677
+ IDnum * cA = (IDnum *) A;
678
+ IDnum * cB = (IDnum *) B;
679
+
680
+ if (*cA > *cB)
681
+ return 1;
682
+ if (*cA == *cB)
683
+ return 0;
684
+ return -1;
685
+ }
686
+
687
+ static void estimateLibraryInsertLength(IDnum * coOccurences, IDnum coOccurencesCount, Category libID) {
688
+ Coordinate median, variance;
689
+ IDnum index;
690
+ int counter = 0;
691
+ qsort(coOccurences, coOccurencesCount, sizeof(IDnum), compareReadOccurences);
692
+
693
+ median = coOccurences[coOccurencesCount / 2];
694
+
695
+ // Modified variance around the median (proxy for expected value)
696
+ // interval censoring
697
+ variance = 0;
698
+ for (index = 0; index < coOccurencesCount; index++) {
699
+ if (coOccurences[index] > 0 && coOccurences[index] < 5 * median) {
700
+ variance += (coOccurences[index] - median) * (coOccurences[index] - median);
701
+ counter++;
702
+ }
703
+ }
704
+ if (counter)
705
+ variance /= counter;
706
+ else {
707
+ variance = 0;
708
+ for (index = 0; index < coOccurencesCount; index++)
709
+ variance += (coOccurences[index] - median) * (coOccurences[index] - median);
710
+ variance /= coOccurencesCount;
711
+ }
712
+
713
+ // To avoid subsequent divisions by zero
714
+ if (variance == 0)
715
+ variance = 1;
716
+
717
+ velvetLog("Paired-end library %i has length: %lli, sample standard deviation: %lli\n", libID + 1, (long long) median, (long long) sqrt(variance));
718
+ setInsertLengths(graph, libID, median, sqrt(variance));
719
+ estimated[libID] = true;
720
+ }
721
+
722
+ static void estimateLibraryInsertLengths(IDnum ** coOccurences, IDnum * coOccurencesCounts) {
723
+ Category libID;
724
+
725
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
726
+ estimated[libID] = false;
727
+
728
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
729
+ if (coOccurencesCounts[libID] > 0)
730
+ estimateLibraryInsertLength(coOccurences[libID], coOccurencesCounts[libID], libID);
731
+ }
732
+
733
+ static void estimateMissingInsertLengths(ReadOccurence ** readNodes, IDnum * readNodeCounts, IDnum * readPairs, Category * cats) {
734
+ IDnum * coOccurences[CATEGORIES + 1];
735
+ IDnum coOccurencesCounts[CATEGORIES + 1];
736
+ Category libID;
737
+
738
+ velvetLog("Estimating library insert lengths...\n");
739
+
740
+ unsigned char * interestingReads = countCoOccurences(coOccurencesCounts, readNodes, readNodeCounts, readPairs, cats);
741
+
742
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
743
+ coOccurences[libID] = callocOrExit(coOccurencesCounts[libID], IDnum);
744
+
745
+ measureCoOccurences(coOccurences, interestingReads, readNodes, readNodeCounts, readPairs, cats);
746
+ estimateLibraryInsertLengths(coOccurences, coOccurencesCounts);
747
+
748
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
749
+ free(coOccurences[libID]);
750
+
751
+ free(interestingReads);
752
+
753
+ velvetLog("Done\n");
754
+ }
755
+
756
+ static void createTwinConnection(IDnum nodeID, IDnum node2ID,
757
+ Connection * connect)
758
+ {
759
+ Connection *newConnection = allocateConnection();
760
+ IDnum nodeIndex = nodeID + nodeCount(graph);
761
+
762
+ // Fill in
763
+ newConnection->distance = connect->distance;
764
+ newConnection->variance = connect->variance;
765
+ newConnection->direct_count = connect->direct_count;
766
+ newConnection->paired_count = connect->paired_count;
767
+ newConnection->destination = getNodeInGraph(graph, node2ID);
768
+
769
+ // Batch to twin
770
+ newConnection->twin = connect;
771
+ connect->twin = newConnection;
772
+
773
+ // Insert in scaffold
774
+ newConnection->left = NULL;
775
+ newConnection->right = scaffold[nodeIndex];
776
+ if (scaffold[nodeIndex] != NULL)
777
+ scaffold[nodeIndex]->left = newConnection;
778
+ scaffold[nodeIndex] = newConnection;
779
+ }
780
+
781
+ Connection *createNewConnection(IDnum nodeID, IDnum node2ID,
782
+ IDnum direct_count,
783
+ IDnum paired_count,
784
+ Coordinate distance,
785
+ double variance)
786
+ {
787
+ Node *destination = getNodeInGraph(graph, node2ID);
788
+ IDnum nodeIndex = nodeID + nodeCount(graph);
789
+ Connection *connect = allocateConnection();
790
+
791
+ // Fill in
792
+ connect->destination = destination;
793
+ connect->direct_count = direct_count;
794
+ connect->paired_count = paired_count;
795
+ connect->distance = (double) distance;
796
+ connect->variance = variance;
797
+
798
+ // Insert in scaffold
799
+ connect->left = NULL;
800
+ connect->right = scaffold[nodeIndex];
801
+ if (scaffold[nodeIndex] != NULL)
802
+ scaffold[nodeIndex]->left = connect;
803
+ scaffold[nodeIndex] = connect;
804
+
805
+ // Event. pair up to twin
806
+ if (getUniqueness(destination))
807
+ createTwinConnection(node2ID, nodeID, connect);
808
+ else
809
+ connect->twin = NULL;
810
+
811
+ return connect;
812
+ }
813
+
814
+ void readjustConnection(Connection * connect, Coordinate distance,
815
+ double variance, IDnum direct_count,
816
+ IDnum paired_count)
817
+ {
818
+ connect->direct_count += direct_count;
819
+ connect->paired_count += paired_count;
820
+
821
+ connect->distance =
822
+ (variance * connect->distance +
823
+ distance * connect->variance) / (variance +
824
+ connect->variance);
825
+ connect->variance =
826
+ (variance *
827
+ connect->variance) / (variance + connect->variance);
828
+
829
+ if (connect->twin != NULL) {
830
+ connect->twin->distance = connect->distance;
831
+ connect->twin->variance = connect->variance;
832
+ connect->twin->direct_count = connect->direct_count;
833
+ connect->twin->paired_count = connect->paired_count;
834
+ }
835
+ }
836
+
837
+ //////////////////////////////////////
838
+ // Splay tree function for Connections
839
+ //////////////////////////////////////
840
+
841
+ /* This function can be called only if K2 has a left child */
842
+ /* Perform a rotate between a node (K2) and its left child */
843
+ /* Update heights, then return new root */
844
+
845
+ static Connection *connectionSingleRotateWithLeft(Connection * K2)
846
+ {
847
+ Connection *K1;
848
+
849
+ K1 = K2->left;
850
+ K2->left = K1->right;
851
+ K1->right = K2;
852
+
853
+ return K1; /* New root */
854
+ }
855
+
856
+ /* This function can be called only if K1 has a right child */
857
+ /* Perform a rotate between a node (K1) and its right child */
858
+ /* Update heights, then return new root */
859
+
860
+ static Connection *connectionSingleRotateWithRight(Connection * K1)
861
+ {
862
+ Connection *K2;
863
+
864
+ K2 = K1->right;
865
+ K1->right = K2->left;
866
+ K2->left = K1;
867
+
868
+ return K2; /* New root */
869
+ }
870
+
871
+ /* Top-down splay procedure, */
872
+ /* not requiring destination to be in tree */
873
+
874
+ static Connection *splayConnection(Connection * T, IDnum nodeID)
875
+ {
876
+ Connection Header;
877
+ Connection *LeftTreeMax, *RightTreeMin;
878
+
879
+ if (T == NULL)
880
+ return NULL;
881
+
882
+ Header.left = Header.right = NULL;
883
+ LeftTreeMax = RightTreeMin = &Header;
884
+
885
+ while (nodeID != getNodeID(T->destination))
886
+ {
887
+ if (nodeID < getNodeID(T->destination))
888
+ {
889
+ if (T->left == NULL)
890
+ break;
891
+ if (nodeID < getNodeID(T->left->destination))
892
+ T = connectionSingleRotateWithLeft(T);
893
+ if (T->left == NULL)
894
+ break;
895
+ /* Link right */
896
+ RightTreeMin->left = T;
897
+ RightTreeMin = T;
898
+ T = T->left;
899
+ }
900
+ else
901
+ {
902
+ if (T->right == NULL)
903
+ break;
904
+ if (nodeID > getNodeID(T->right->destination))
905
+ T = connectionSingleRotateWithRight(T);
906
+ if (T->right == NULL)
907
+ break;
908
+ /* Link left */
909
+ LeftTreeMax->right = T;
910
+ LeftTreeMax = T;
911
+ T = T->right;
912
+ }
913
+ } /* while nodeID != T->destination */
914
+
915
+ /* Reassemble */
916
+ LeftTreeMax->right = T->left;
917
+ RightTreeMin->left = T->right;
918
+ T->left = Header.right;
919
+ T->right = Header.left;
920
+
921
+ return T;
922
+ }
923
+
924
+ static Connection* findOrCreateConnection(IDnum nodeID,
925
+ IDnum node2ID)
926
+ {
927
+ Connection **T;
928
+ Connection *newConnection;
929
+ IDnum nodeIndex;
930
+
931
+ nodeIndex = nodeID + nodeCount(graph);
932
+ T = scaffold + nodeIndex;
933
+
934
+ if (*T == NULL)
935
+ {
936
+ newConnection = allocateConnection();
937
+
938
+ newConnection->left = NULL;
939
+ newConnection->right = NULL;
940
+ *T = newConnection;
941
+ }
942
+ else
943
+ {
944
+ IDnum destID;
945
+
946
+ *T = splayConnection(*T, node2ID);
947
+ destID = getNodeID((*T)->destination);
948
+ if (destID == node2ID)
949
+ newConnection = *T;
950
+ else
951
+ {
952
+ newConnection = allocateConnection();
953
+ if (node2ID < destID)
954
+ {
955
+ newConnection->left = (*T)->left;
956
+ newConnection->right = *T;
957
+ (*T)->left = NULL;
958
+ }
959
+ else if (node2ID > destID)
960
+ {
961
+ newConnection->right = (*T)->right;
962
+ newConnection->left = *T;
963
+ (*T)->right = NULL;
964
+ }
965
+ *T = newConnection;
966
+ }
967
+ }
968
+
969
+ return newConnection;
970
+ }
971
+
972
+ static Connection* findConnection(IDnum nodeID,
973
+ IDnum node2ID)
974
+ {
975
+ Connection **T;
976
+ IDnum nodeIndex;
977
+
978
+ nodeIndex = nodeID + nodeCount(graph);
979
+ T = scaffold + nodeIndex;
980
+
981
+ if (*T == NULL)
982
+ return NULL;
983
+ else
984
+ {
985
+ IDnum destID;
986
+
987
+ *T = splayConnection(*T, node2ID);
988
+ destID = getNodeID((*T)->destination);
989
+ if (destID == node2ID)
990
+ return *T;
991
+ }
992
+ return NULL;
993
+ }
994
+
995
+ RecycleBin *connectionStackMemory = NULL;
996
+
997
+ typedef struct ConnectionStack_st ConnectionStack;
998
+
999
+ struct ConnectionStack_st
1000
+ {
1001
+ Connection *connection;
1002
+ ConnectionStack *next;
1003
+ };
1004
+
1005
+ #ifdef _OPENMP
1006
+ static void initConnectionStackMemory(void)
1007
+ {
1008
+ int n = omp_get_max_threads();
1009
+
1010
+ #pragma omp critical
1011
+ {
1012
+ if (connectionStackMemory == NULL)
1013
+ connectionStackMemory = newRecycleBinArray(n, sizeof(ConnectionStack), BLOCK_SIZE);
1014
+ }
1015
+ }
1016
+ #endif
1017
+
1018
+ static ConnectionStack *allocateConnectionStack(void)
1019
+ {
1020
+ #ifdef _OPENMP
1021
+ #ifdef DEBUG
1022
+ if (connectionStackMemory == NULL)
1023
+ {
1024
+ velvetLog("The memory for connection stack seems uninitialised, "
1025
+ "this is probably a bug, aborting.\n");
1026
+ abort();
1027
+ }
1028
+ #endif
1029
+ return allocatePointer(getRecycleBinInArray(connectionStackMemory,
1030
+ omp_get_thread_num()));
1031
+ #else
1032
+ if (connectionStackMemory == NULL)
1033
+ connectionStackMemory =
1034
+ newRecycleBin(sizeof(ConnectionStack), BLOCK_SIZE);
1035
+
1036
+ return allocatePointer(connectionStackMemory);
1037
+ #endif
1038
+ }
1039
+
1040
+ static void deallocateConnectionStack(ConnectionStack *stack)
1041
+ {
1042
+ #ifdef _OPENMP
1043
+ deallocatePointer(getRecycleBinInArray(connectionStackMemory,
1044
+ omp_get_thread_num()),
1045
+ stack);
1046
+ #else
1047
+ deallocatePointer(connectionStackMemory, stack);
1048
+ #endif
1049
+ }
1050
+
1051
+ static void destroyConnectionStackMemory(void)
1052
+ {
1053
+ #ifdef _OPENMP
1054
+ destroyRecycleBinArray(connectionStackMemory);
1055
+ #else
1056
+ destroyRecycleBin(connectionStackMemory);
1057
+ #endif
1058
+ connectionStackMemory = NULL;
1059
+ }
1060
+
1061
+ static void pushConnectionStack(ConnectionStack **stack, Connection *connection)
1062
+ {
1063
+ ConnectionStack *newElement;
1064
+
1065
+ newElement = allocateConnectionStack();
1066
+ newElement->connection = connection;
1067
+ newElement->next = *stack;
1068
+ *stack = newElement;
1069
+ }
1070
+
1071
+ static Connection *popConnectionStack(ConnectionStack **stack)
1072
+ {
1073
+ ConnectionStack *nextElement;
1074
+ Connection *connection;
1075
+
1076
+ if (*stack == NULL)
1077
+ return NULL;
1078
+
1079
+ nextElement = (*stack)->next;
1080
+ connection = (*stack)->connection;
1081
+ deallocateConnectionStack(*stack);
1082
+ *stack = nextElement;
1083
+
1084
+ return connection;
1085
+ }
1086
+
1087
+ static void splayToList(Connection **connection)
1088
+ {
1089
+ ConnectionStack *stack = NULL;
1090
+ Connection *current;
1091
+ Connection *list = NULL;
1092
+
1093
+ if (*connection == NULL)
1094
+ return;
1095
+
1096
+ for (current = *connection; current != NULL; current = popConnectionStack(&stack))
1097
+ {
1098
+ Connection *right;
1099
+ Connection *left;
1100
+
1101
+ right = current->right;
1102
+ if (right != NULL)
1103
+ pushConnectionStack(&stack, right);
1104
+ left = current->left;
1105
+ if (left != NULL)
1106
+ pushConnectionStack(&stack, left);
1107
+ if (list != NULL)
1108
+ list->left = current;
1109
+ current->right = list;
1110
+ list = current;
1111
+ }
1112
+ list->left = NULL;
1113
+ *connection = list;
1114
+ }
1115
+
1116
+ static void setAllConnectionsClean(void)
1117
+ {
1118
+ IDnum nodeID;
1119
+ IDnum nodes = nodeCount(graph);
1120
+
1121
+ #ifdef _OPENMP
1122
+ #pragma omp parallel for
1123
+ #endif
1124
+ for (nodeID = 2 * nodes; nodeID >= 0; nodeID--)
1125
+ {
1126
+ ConnectionStack *stack = NULL;
1127
+ Connection **connect;
1128
+ Connection *current;
1129
+
1130
+ connect = scaffold + nodeID;
1131
+ if (*connect == NULL)
1132
+ continue;
1133
+
1134
+ for (current = *connect; current != NULL; current = popConnectionStack(&stack))
1135
+ {
1136
+ Connection *right;
1137
+ Connection *left;
1138
+
1139
+ current->clean = true;
1140
+ right = current->right;
1141
+ if (right != NULL)
1142
+ pushConnectionStack(&stack, right);
1143
+ left = current->left;
1144
+ if (left != NULL)
1145
+ pushConnectionStack(&stack, left);
1146
+ }
1147
+ }
1148
+ }
1149
+
1150
+ static void fillNewConnectionInTree(Connection *connect,
1151
+ Node *destination,
1152
+ IDnum direct_count,
1153
+ IDnum paired_count,
1154
+ Coordinate distance,
1155
+ double variance)
1156
+ {
1157
+ connect->destination = destination;
1158
+ connect->direct_count = direct_count;
1159
+ connect->paired_count = paired_count;
1160
+ connect->distance = (double)distance;
1161
+ connect->variance = variance;
1162
+ }
1163
+
1164
+ static void readjustConnectionInTree(Connection *connect,
1165
+ IDnum direct_count,
1166
+ IDnum paired_count,
1167
+ Coordinate distance,
1168
+ double variance)
1169
+ {
1170
+ connect->direct_count += direct_count;
1171
+ connect->paired_count += paired_count;
1172
+ connect->distance = (variance * connect->distance + distance * connect->variance) /
1173
+ (variance + connect->variance);
1174
+ connect->variance = (variance * connect->variance) / (variance + connect->variance);
1175
+
1176
+ if (connect->twin != NULL)
1177
+ {
1178
+ connect->twin->direct_count = connect->direct_count;
1179
+ connect->twin->paired_count = connect->paired_count;
1180
+ connect->twin->distance = connect->distance;
1181
+ connect->twin->variance = connect->variance;
1182
+ }
1183
+ }
1184
+
1185
+ static void createTwinConnectionInTree(IDnum nodeID,
1186
+ IDnum node2ID,
1187
+ Connection *connect)
1188
+ {
1189
+ Connection *newConnection;
1190
+
1191
+ newConnection = findOrCreateConnection(nodeID, node2ID);
1192
+ if (newConnection->destination == NULL)
1193
+ {
1194
+ fillNewConnectionInTree(newConnection,
1195
+ getNodeInGraph(graph, node2ID),
1196
+ connect->direct_count,
1197
+ connect->paired_count,
1198
+ (Coordinate)connect->distance,
1199
+ connect->variance);
1200
+ // Batch to twin
1201
+ newConnection->twin = connect;
1202
+ connect->twin = newConnection;
1203
+ }
1204
+ else
1205
+ readjustConnectionInTree(newConnection,
1206
+ connect->direct_count,
1207
+ connect->paired_count,
1208
+ (Coordinate)connect->distance,
1209
+ connect->variance);
1210
+ }
1211
+
1212
+ static void createConnection(IDnum nodeID,
1213
+ IDnum node2ID,
1214
+ IDnum direct_count,
1215
+ IDnum paired_count,
1216
+ Coordinate distance,
1217
+ double variance)
1218
+ {
1219
+ Connection *connect;
1220
+
1221
+ if (getUniqueness(getNodeInGraph(graph, node2ID)) && node2ID < nodeID) {
1222
+ return;
1223
+ }
1224
+
1225
+ #ifdef _OPENMP
1226
+ lockTwoNodes(nodeID, node2ID);
1227
+ #endif
1228
+ connect = findOrCreateConnection(nodeID, node2ID);
1229
+ if (connect->destination == NULL)
1230
+ {
1231
+ Node *destination = getNodeInGraph(graph, node2ID);
1232
+ fillNewConnectionInTree(connect,
1233
+ destination,
1234
+ direct_count,
1235
+ paired_count,
1236
+ distance,
1237
+ variance);
1238
+
1239
+ if (getUniqueness(destination))
1240
+ createTwinConnectionInTree(node2ID, nodeID, connect);
1241
+ else
1242
+ connect->twin = NULL;
1243
+ }
1244
+ else
1245
+ readjustConnectionInTree(connect,
1246
+ direct_count,
1247
+ paired_count,
1248
+ distance,
1249
+ variance);
1250
+
1251
+ #ifdef _OPENMP
1252
+ unLockTwoNodes(nodeID, node2ID);
1253
+ #endif
1254
+ }
1255
+
1256
+ static void projectFromSingleRead(Node * node,
1257
+ ReadOccurence * readOccurence,
1258
+ Coordinate position,
1259
+ Coordinate offset, Coordinate length)
1260
+ {
1261
+ Coordinate distance = 0;
1262
+ Node *target = getNodeInGraph(graph, -readOccurence->nodeID);
1263
+ double variance = 1;
1264
+
1265
+ if (target == getTwinNode(node) || target == node)
1266
+ return;
1267
+
1268
+ if (position < 0) {
1269
+ variance += getNodeLength(node) * getNodeLength(node) / 16;
1270
+ // distance += 0;
1271
+ } else {
1272
+ // variance += 0;
1273
+ distance += position - getNodeLength(node) / 2;
1274
+ }
1275
+
1276
+ if (readOccurence->position < 0) {
1277
+ variance +=
1278
+ getNodeLength(target) * getNodeLength(target) / 16;
1279
+ //distance += 0;
1280
+ } else {
1281
+ // variance += 0;
1282
+ distance +=
1283
+ -readOccurence->position + getNodeLength(target) / 2;
1284
+ }
1285
+
1286
+ if (readOccurence->offset < 0 || offset < 0) {
1287
+ variance += length * length / 16;
1288
+ //distance += 0;
1289
+ } else {
1290
+ // variance += 0;
1291
+ distance += readOccurence->offset - offset;
1292
+ }
1293
+
1294
+ // Relative ordering
1295
+ if (offset > 0 && readOccurence->offset > 0) {
1296
+ if (offset < readOccurence->offset) {
1297
+ if (distance - getNodeLength(node)/2 - getNodeLength(target)/2 < -10)
1298
+ ;
1299
+ else if (distance < getNodeLength(node)/2 + getNodeLength(target)/2)
1300
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1301
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1302
+ else
1303
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1304
+ distance, variance);
1305
+ } else if (offset > readOccurence->offset) {
1306
+ if (-distance - getNodeLength(node)/2 - getNodeLength(target)/2 < -10)
1307
+ ;
1308
+ else if (-distance < getNodeLength(node)/2 + getNodeLength(target)/2)
1309
+ createConnection(-getNodeID(node), -getNodeID(target), 1,
1310
+ 0, getNodeLength(node)/2 + getNodeLength(target)/2 , variance);
1311
+ else
1312
+ createConnection(-getNodeID(node), -getNodeID(target), 1,
1313
+ 0, -distance, variance);
1314
+ }
1315
+ } else if (offset > 0 && position > 0) {
1316
+ if (distance - offset > -getNodeLength(node)/2 && distance - offset + length > getNodeLength(node)/2)
1317
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1318
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1319
+ else if (distance - offset < -getNodeLength(node)/2 && distance - offset + length < getNodeLength(node)/2)
1320
+ createConnection(-getNodeID(node), -getNodeID(target), 1, 0,
1321
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1322
+ else {
1323
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1324
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1325
+ createConnection(-getNodeID(node), -getNodeID(target), 1, 0,
1326
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1327
+ }
1328
+ } else if (readOccurence->offset > 0 && readOccurence->position > 0) {
1329
+ if (-distance - readOccurence->offset > -getNodeLength(target)/2 && -distance - readOccurence->offset + length > getNodeLength(target)/2)
1330
+ createConnection(-getNodeID(node), -getNodeID(target), 1, 0,
1331
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1332
+ if (-distance - readOccurence->offset < -getNodeLength(target)/2 && -distance - readOccurence->offset + length < getNodeLength(target)/2)
1333
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1334
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1335
+ else {
1336
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1337
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1338
+ createConnection(-getNodeID(node), -getNodeID(target), 1, 0,
1339
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1340
+ }
1341
+ } else {
1342
+ createConnection(getNodeID(node), getNodeID(target), 1, 0,
1343
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1344
+ createConnection(-getNodeID(node), -getNodeID(target), 1, 0,
1345
+ getNodeLength(node)/2 + getNodeLength(target)/2, variance);
1346
+ }
1347
+ }
1348
+
1349
+ static void projectFromReadPair(Node * node, ReadOccurence * readOccurence,
1350
+ Coordinate position, Coordinate offset,
1351
+ Coordinate insertLength,
1352
+ double insertVariance,
1353
+ boolean doMatePairs)
1354
+ {
1355
+ Coordinate distance = insertLength;
1356
+ Coordinate variance = insertVariance;
1357
+ Node *target = getNodeInGraph(graph, readOccurence->nodeID);
1358
+ IDnum nodeID;
1359
+ IDnum node2ID;
1360
+
1361
+ if (target == getTwinNode(node) || target == node)
1362
+ return;
1363
+
1364
+ nodeID = getNodeID(node);
1365
+ node2ID = getNodeID(target);
1366
+
1367
+ if (getUniqueness(target) && node2ID < nodeID)
1368
+ return;
1369
+
1370
+ // Check if a conflicting PE (or MP from a smaller size lib) connection
1371
+ // already exists
1372
+ if (doMatePairs) {
1373
+ Connection *reverseConnect;
1374
+
1375
+ #ifdef _OPENMP
1376
+ lockTwoNodes(nodeID, node2ID);
1377
+ #endif
1378
+ reverseConnect = findConnection(-nodeID, -node2ID);
1379
+ #ifdef _OPENMP
1380
+ unLockTwoNodes(nodeID, node2ID);
1381
+ #endif
1382
+
1383
+ if (reverseConnect != NULL &&
1384
+ reverseConnect->clean &&
1385
+ reverseConnect->paired_count +
1386
+ reverseConnect->direct_count >= UNRELIABLE_CONNECTION_CUTOFF)
1387
+ return;
1388
+ }
1389
+
1390
+ if (position < 0) {
1391
+ variance += getNodeLength(node) * getNodeLength(node) / 16;
1392
+ // distance += 0;
1393
+ } else {
1394
+ // variance += 0;
1395
+ distance += position - offset - getNodeLength(node) / 2;
1396
+ }
1397
+
1398
+ if (readOccurence->position < 0) {
1399
+ variance +=
1400
+ getNodeLength(target) * getNodeLength(target) / 16;
1401
+ //distance += 0;
1402
+ } else {
1403
+ // variance += 0;
1404
+ distance +=
1405
+ readOccurence->position - readOccurence->offset -
1406
+ getNodeLength(target) / 2;
1407
+ }
1408
+
1409
+ if (distance - getNodeLength(node)/2 - getNodeLength(target)/2 < -6 * sqrt(insertVariance))
1410
+ return;
1411
+ else if (distance < getNodeLength(node)/2 + getNodeLength(target)/2)
1412
+ distance = getNodeLength(node)/2 + getNodeLength(target)/2;
1413
+
1414
+ createConnection(nodeID, node2ID, 0, 1,
1415
+ distance, variance);
1416
+ }
1417
+
1418
+ static void projectFromShortRead(Node * node,
1419
+ ShortReadMarker * shortMarker,
1420
+ IDnum * readPairs, Category * cats,
1421
+ ReadOccurence ** readNodes,
1422
+ IDnum * readNodeCounts,
1423
+ ShortLength * lengths,
1424
+ boolean * shadows,
1425
+ boolean doMatePairs,
1426
+ Category thisCat)
1427
+ {
1428
+ IDnum index;
1429
+ IDnum readIndex = getShortReadMarkerID(shortMarker);
1430
+ ReadOccurence *readArray;
1431
+ IDnum readPairIndex;
1432
+ Category cat;
1433
+ Coordinate position = getShortReadMarkerPosition(shortMarker);
1434
+ Coordinate offset = getShortReadMarkerOffset(shortMarker);
1435
+ Coordinate length = lengths[getShortReadMarkerID(shortMarker) - 1];
1436
+ Coordinate insertLength;
1437
+ double insertVariance;
1438
+
1439
+ // Going through single-read information
1440
+ if (!doMatePairs && readNodeCounts[readIndex] > 1) {
1441
+ readArray = readNodes[readIndex];
1442
+ for (index = 0; index < readNodeCounts[readIndex]; index++)
1443
+ projectFromSingleRead(node, &readArray[index],
1444
+ position, offset, length);
1445
+ }
1446
+ // Going through paired read information
1447
+ if (readPairs == NULL)
1448
+ return;
1449
+
1450
+ readPairIndex = readPairs[readIndex - 1] + 1;
1451
+
1452
+ if (readPairIndex == 0)
1453
+ return;
1454
+
1455
+ cat = cats[readIndex - 1];
1456
+ insertLength = getInsertLength(graph, cat);
1457
+ insertVariance = getInsertLength_var(graph, cat);
1458
+ cat /= 2;
1459
+ if (shadows[cat] && cat > PEBBLE_ROUND_NUM)
1460
+ return;
1461
+
1462
+ if (!shadows[cat] && !doMatePairs) {
1463
+ readArray = readNodes[readPairIndex];
1464
+ for (index = 0; index < readNodeCounts[readPairIndex]; index++)
1465
+ projectFromReadPair(node, &readArray[index], position,
1466
+ offset, insertLength, insertVariance, false);
1467
+ }
1468
+ else if (shadows[cat] && doMatePairs && cat == thisCat) {
1469
+ readArray = readNodes[readPairIndex];
1470
+ for (index = 0; index < readNodeCounts[readPairIndex]; index++)
1471
+ projectFromReadPair(node, &readArray[index], position,
1472
+ offset, insertLength, insertVariance, true);
1473
+ }
1474
+
1475
+ }
1476
+
1477
+ static void projectFromLongRead(Node * node, PassageMarkerI marker,
1478
+ IDnum * readPairs, Category * cats,
1479
+ ReadOccurence ** readNodes,
1480
+ IDnum * readNodeCounts,
1481
+ ShortLength * lengths)
1482
+ {
1483
+ IDnum index;
1484
+ IDnum readIndex = getPassageMarkerSequenceID(marker);
1485
+ ReadOccurence *readArray;
1486
+ IDnum readPairIndex;
1487
+ Category cat;
1488
+ Coordinate position = getStartOffset(marker);
1489
+ Coordinate offset = getPassageMarkerStart(marker);
1490
+ Coordinate length =
1491
+ lengths[getPassageMarkerSequenceID(marker) - 1];
1492
+ Coordinate insertLength;
1493
+ double insertVariance;
1494
+
1495
+ // Going through single-read information
1496
+ if (readNodeCounts[readIndex] > 1 && position > 0) {
1497
+ readArray = readNodes[readIndex];
1498
+ for (index = 0; index < readNodeCounts[readIndex]; index++)
1499
+ projectFromSingleRead(node, &readArray[index],
1500
+ position, offset, length);
1501
+ }
1502
+ // Going through paired read information
1503
+ if (readPairs == NULL)
1504
+ return;
1505
+
1506
+ readPairIndex = readPairs[readIndex - 1] + 1;
1507
+
1508
+ if (readPairIndex == 0)
1509
+ return;
1510
+
1511
+ cat = cats[readIndex - 1];
1512
+ insertLength = getInsertLength(graph, cat);
1513
+ insertVariance = getInsertLength_var(graph, cat);
1514
+
1515
+ readArray = readNodes[readPairIndex];
1516
+ for (index = 0; index < readNodeCounts[readPairIndex]; index++)
1517
+ projectFromReadPair(node, &readArray[index], position,
1518
+ offset, insertLength, insertVariance, false);
1519
+
1520
+ }
1521
+
1522
+ static void projectFromNode(IDnum nodeID,
1523
+ ReadOccurence ** readNodes,
1524
+ IDnum * readNodeCounts,
1525
+ IDnum * readPairs, Category * cats,
1526
+ boolean * dubious, ShortLength * lengths,
1527
+ boolean * shadows,
1528
+ boolean doMatePairs,
1529
+ Category thisCat)
1530
+ {
1531
+ IDnum index;
1532
+ ShortReadMarker *nodeArray, *shortMarker;
1533
+ PassageMarkerI marker;
1534
+ Node *node;
1535
+ IDnum nodeReadCount;
1536
+
1537
+ node = getNodeInGraph(graph, nodeID);
1538
+
1539
+ if (node == NULL || !getUniqueness(node))
1540
+ return;
1541
+
1542
+ nodeArray = getNodeReads(node, graph);
1543
+ nodeReadCount = getNodeReadCount(node, graph);
1544
+ for (index = 0; index < nodeReadCount; index++) {
1545
+ shortMarker = getShortReadMarkerAtIndex(nodeArray, index);
1546
+ if (dubious[getShortReadMarkerID(shortMarker) - 1])
1547
+ continue;
1548
+ projectFromShortRead(node, shortMarker, readPairs, cats,
1549
+ readNodes, readNodeCounts, lengths,
1550
+ shadows,
1551
+ doMatePairs,
1552
+ thisCat);
1553
+ }
1554
+
1555
+ if (!doMatePairs)
1556
+ for (marker = getMarker(node); marker != NULL_IDX;
1557
+ marker = getNextInNode(marker)) {
1558
+ if (getPassageMarkerSequenceID(marker) > 0)
1559
+ projectFromLongRead(node, marker, readPairs, cats,
1560
+ readNodes, readNodeCounts,
1561
+ lengths);
1562
+ }
1563
+ }
1564
+
1565
+ static Connection **computeNodeToNodeMappings(ReadOccurence ** readNodes,
1566
+ IDnum * readNodeCounts,
1567
+ IDnum * readPairs,
1568
+ Category * cats,
1569
+ boolean * dubious,
1570
+ boolean * shadows,
1571
+ ShortLength * lengths)
1572
+ {
1573
+ IDnum nodeID;
1574
+ IDnum nodes = nodeCount(graph);
1575
+ struct timeval start, end, diff;
1576
+ Category cat;
1577
+ boolean hasShadow;
1578
+
1579
+ scaffold = callocOrExit(2 * nodes + 1, Connection *);
1580
+
1581
+ velvetLog("Computing direct node to node mappings\n");
1582
+
1583
+ gettimeofday(&start, NULL);
1584
+ #ifdef _OPENMP
1585
+ createNodeLocks(graph);
1586
+
1587
+ int threads = omp_get_max_threads();
1588
+ if (threads > 32)
1589
+ threads = 32;
1590
+
1591
+ #pragma omp parallel for num_threads(threads)
1592
+ #endif
1593
+ for (nodeID = -nodes; nodeID <= nodes; nodeID++)
1594
+ {
1595
+ if (nodeID % 10000 == 0)
1596
+ velvetLog("Scaffolding node %li\n", (long) nodeID);
1597
+
1598
+ projectFromNode(nodeID, readNodes, readNodeCounts,
1599
+ readPairs, cats, dubious, lengths, shadows, false, 0);
1600
+ }
1601
+
1602
+ #ifdef _OPENMP
1603
+ initConnectionStackMemory();
1604
+ #endif
1605
+
1606
+ hasShadow = false;
1607
+ for (cat = 0; cat < CATEGORIES; cat++)
1608
+ if (shadows[cat])
1609
+ {
1610
+ hasShadow = true;
1611
+ break;
1612
+ }
1613
+
1614
+ if (hasShadow)
1615
+ {
1616
+ for (cat = 0; cat < CATEGORIES; cat++)
1617
+ {
1618
+ setAllConnectionsClean();
1619
+ if (!shadows[cat])
1620
+ continue;
1621
+ velvetLog("Scaffolding MP library %i\n", cat);
1622
+ #ifdef _OPENMP
1623
+ #pragma omp parallel for
1624
+ #endif
1625
+ for (nodeID = -nodes; nodeID <= nodes; nodeID++)
1626
+ projectFromNode(nodeID, readNodes, readNodeCounts,
1627
+ readPairs, cats, dubious, lengths,
1628
+ shadows, true, cat);
1629
+ }
1630
+ }
1631
+ #ifdef _OPENMP
1632
+ #pragma omp parallel for
1633
+ #endif
1634
+ for (nodeID = 2 * nodes; nodeID >= 0; nodeID--)
1635
+ splayToList(scaffold + nodeID);
1636
+
1637
+ destroyConnectionStackMemory();
1638
+
1639
+ #ifdef _OPENMP
1640
+ free(nodeLocks);
1641
+ nodeLocks = NULL;
1642
+ #endif
1643
+ gettimeofday(&end, NULL);
1644
+ timersub(&end, &start, &diff);
1645
+ velvetLog(" === Nodes Scaffolded in %ld.%06ld s\n", (long) diff.tv_sec, (long) diff.tv_usec);
1646
+
1647
+ PEBBLE_ROUND_NUM++;
1648
+
1649
+ return scaffold;
1650
+ }
1651
+
1652
+ static IDnum **countShortReads(Graph * graph, ReadSet * reads)
1653
+ {
1654
+ IDnum **counts = callocOrExit(CATEGORIES + 1, IDnum *);
1655
+ Category cat;
1656
+ IDnum nodeIndex;
1657
+ IDnum nodes = nodeCount(graph);
1658
+ Node *node;
1659
+ ShortReadMarker *array, *marker;
1660
+ IDnum readCount, readIndex, readID;
1661
+
1662
+ // Allocate memory where needed
1663
+ for (cat = 0; cat <= CATEGORIES; cat++)
1664
+ if (getInsertLength(graph, cat) > 0)
1665
+ counts[cat] =
1666
+ callocOrExit(2 * nodeCount(graph) + 1,
1667
+ IDnum);
1668
+
1669
+ // Start fillin'
1670
+ for (nodeIndex = 0; nodeIndex < 2 * nodes + 1; nodeIndex++) {
1671
+ node = getNodeInGraph(graph, nodeIndex - nodes);
1672
+
1673
+ if (node == NULL || !getUniqueness(node))
1674
+ continue;
1675
+
1676
+ array = getNodeReads(node, graph);
1677
+ readCount = getNodeReadCount(node, graph);
1678
+ for (readIndex = 0; readIndex < readCount; readIndex++) {
1679
+ marker =
1680
+ getShortReadMarkerAtIndex(array, readIndex);
1681
+ readID = getShortReadMarkerID(marker);
1682
+ cat = reads->categories[readID - 1];
1683
+ if (cat % 2 == 1 && counts[cat / 2] != NULL)
1684
+ counts[cat / 2][nodeIndex]++;
1685
+ }
1686
+ }
1687
+
1688
+ return counts;
1689
+ }
1690
+
1691
+ static void removeUnreliableConnections(ReadSet * reads, boolean *shadows)
1692
+ {
1693
+ IDnum maxNodeIndex = nodeCount(graph) * 2 + 1;
1694
+ IDnum index;
1695
+ Connection *connect, *next;
1696
+ Category cat;
1697
+ IDnum **counts = countShortReads(graph, reads);
1698
+ IDnum nodes = nodeCount(graph);
1699
+
1700
+ for (index = 0; index < maxNodeIndex; index++) {
1701
+ for (connect = scaffold[index]; connect != NULL;
1702
+ connect = next) {
1703
+ next = connect->right;
1704
+ if (!testConnection(index - nodes, connect, counts, shadows))
1705
+ destroyConnection(connect, index - nodes);
1706
+ }
1707
+ }
1708
+
1709
+ // Free memory
1710
+ for (cat = 0; cat <= CATEGORIES; cat++)
1711
+ if (counts[cat])
1712
+ free(counts[cat]);
1713
+ free(counts);
1714
+ }
1715
+
1716
+ void printConnections(ReadSet * reads, boolean * shadows)
1717
+ {
1718
+ IDnum maxNodeIndex = nodeCount(graph) * 2 + 1;
1719
+ IDnum index;
1720
+ Connection *connect, *next;
1721
+ Node *node;
1722
+ IDnum **counts = countShortReads(graph, reads);
1723
+ IDnum nodes = nodeCount(graph);
1724
+ Category cat;
1725
+
1726
+ puts("CONNECT IDA IDB dcount pcount dist lengthA lengthB var countA countB coordA coordB real exp distance test");
1727
+
1728
+ for (index = 0; index < maxNodeIndex; index++) {
1729
+ node = getNodeInGraph(graph, index - nodeCount(graph));
1730
+ for (connect = scaffold[index]; connect != NULL;
1731
+ connect = next) {
1732
+ next = getNextConnection(connect);
1733
+ printf
1734
+ ("CONNECT %ld %ld %ld %ld %lld %lld %lld %f %ld %ld",
1735
+ (long) index - nodeCount(graph),
1736
+ (long) getNodeID(connect->destination),
1737
+ (long) connect->direct_count,
1738
+ (long) connect->paired_count,
1739
+ (long long) getConnectionDistance(connect),
1740
+ (long long) getNodeLength(node), (long long)
1741
+ getNodeLength(connect->destination),
1742
+ connect->variance,
1743
+ (long) getNodeReadCount(node, graph),
1744
+ (long) getNodeReadCount(connect->destination,
1745
+ graph));
1746
+ if (markerCount(node) == 1
1747
+ && markerCount(connect->destination) == 1)
1748
+ printf(" %lld %lld %lld", (long long)
1749
+ getPassageMarkerFinish(getMarker
1750
+ (node)),
1751
+ (long long)
1752
+ getPassageMarkerFinish(getMarker
1753
+ (connect->
1754
+ destination)),
1755
+ (long
1756
+ long) (getPassageMarkerFinish
1757
+ (getMarker(node)) -
1758
+ getPassageMarkerFinish
1759
+ (getMarker
1760
+ (connect->destination))));
1761
+ else
1762
+ printf(" ? ? ?");
1763
+ printf(" %ld",
1764
+ (long) expectedNumberOfConnections(index -
1765
+ nodeCount
1766
+ (graph),
1767
+ connect,
1768
+ counts,
1769
+ 0));
1770
+ printf(" %lld",
1771
+ (long long) (getConnectionDistance(connect)
1772
+ - (getNodeLength(node) +
1773
+ getNodeLength
1774
+ (connect->destination)) /
1775
+ 2));
1776
+ if (testConnection(index - nodes, connect, counts, shadows))
1777
+ puts(" OK");
1778
+ else
1779
+ puts(" NG");
1780
+ }
1781
+ }
1782
+
1783
+ for (cat = 0; cat <= CATEGORIES; cat++)
1784
+ if (counts[cat])
1785
+ free(counts[cat]);
1786
+ free(counts);
1787
+ }
1788
+
1789
+ void buildScaffold(Graph * argGraph,
1790
+ ReadSet * reads,
1791
+ boolean * dubious,
1792
+ boolean * shadows)
1793
+ {
1794
+ IDnum *readPairs;
1795
+ Category *cats;
1796
+ IDnum *readNodeCounts;
1797
+ ReadOccurence **readNodes;
1798
+ ReadOccurence *readNodesArray = NULL;
1799
+ ShortLength *lengths = getSequenceLengths(reads, getWordLength(argGraph));
1800
+ Coordinate totalCount = 0;
1801
+
1802
+ graph = argGraph;
1803
+ readPairs = reads->mateReads;
1804
+ cats = reads->categories;
1805
+
1806
+ // Prepare primary scaffold
1807
+ readNodeCounts = computeReadToNodeCounts(&totalCount);
1808
+ readNodes = computeReadToNodeMappings(readNodeCounts, reads, totalCount, &readNodesArray);
1809
+
1810
+ estimateMissingInsertLengths(readNodes, readNodeCounts, readPairs, cats);
1811
+
1812
+ scaffold = computeNodeToNodeMappings(readNodes, readNodeCounts,
1813
+ readPairs, cats, dubious, shadows, lengths);
1814
+ removeUnreliableConnections(reads, shadows);
1815
+
1816
+ free(readNodesArray);
1817
+ free(readNodes);
1818
+ free(readNodeCounts);
1819
+ free(lengths);
1820
+ }
1821
+
1822
+ //DEBUG
1823
+ void printScaffold(Graph * argGraph,
1824
+ ReadSet * reads,
1825
+ boolean * dubious,
1826
+ boolean * shadows)
1827
+ {
1828
+ IDnum *readPairs;
1829
+ Category *cats;
1830
+ IDnum *readNodeCounts;
1831
+ ReadOccurence **readNodes;
1832
+ ReadOccurence *readNodesArray = NULL;
1833
+ ShortLength *lengths = getSequenceLengths(reads, getWordLength(argGraph));
1834
+ Coordinate totalCount = 0;
1835
+
1836
+ graph = argGraph;
1837
+ readPairs = reads->mateReads;
1838
+ cats = reads->categories;
1839
+
1840
+ // Prepare primary scaffold
1841
+ readNodeCounts = computeReadToNodeCounts(&totalCount);
1842
+ readNodes = computeReadToNodeMappings(readNodeCounts, reads, totalCount, &readNodesArray);
1843
+
1844
+ estimateMissingInsertLengths(readNodes, readNodeCounts, readPairs, cats);
1845
+
1846
+ scaffold = computeNodeToNodeMappings(readNodes, readNodeCounts,
1847
+ readPairs, cats, dubious, shadows, lengths);
1848
+ printConnections(reads, shadows);
1849
+
1850
+ free(readNodesArray);
1851
+ free(readNodes);
1852
+ free(readNodeCounts);
1853
+ free(lengths);
1854
+ cleanScaffoldMemory();
1855
+ }
1856
+
1857
+ void setUnreliableConnectionCutoff(int val)
1858
+ {
1859
+ UNRELIABLE_CONNECTION_CUTOFF = (IDnum) val;
1860
+ }
1861
+
1862
+ void cleanScaffoldMemory() {
1863
+ Category libID;
1864
+
1865
+ for (libID = 0; libID < CATEGORIES + 1; libID++)
1866
+ if (estimated[libID])
1867
+ setInsertLengths(graph, libID, -1, -1);
1868
+
1869
+ destroyRecycleBin(connectionMemory);
1870
+ free(scaffold);
1871
+ connectionMemory = NULL;
1872
+ }
1873
+
1874
+ void setPairedExpFraction(double x) {
1875
+ paired_exp_fraction = x;
1876
+ }