BammetricsReport.scala 21.3 KB
Newer Older
Peter van 't Hof's avatar
Peter van 't Hof committed
1
/**
2 3 4 5 6 7 8 9 10 11 12 13 14
  * Biopet is built on top of GATK Queue for building bioinformatic
  * pipelines. It is mainly intended to support LUMC SHARK cluster which is running
  * SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
  * should also be able to execute Biopet tools and pipelines.
  *
  * Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
  *
  * Contact us at: sasc@lumc.nl
  *
  * A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
  * license; For commercial users or users who do not want to follow the AGPL
  * license, please contact us to obtain a separate license.
  */
15 16
package nl.lumc.sasc.biopet.pipelines.bammetrics

17
import java.io.{File, PrintWriter}
18

Peter van 't Hof's avatar
Peter van 't Hof committed
19
import nl.lumc.sasc.biopet.utils.config.Configurable
20 21 22 23 24 25 26
import nl.lumc.sasc.biopet.core.report.{
  ReportBuilder,
  ReportBuilderExtension,
  ReportPage,
  ReportSection
}
import nl.lumc.sasc.biopet.pipelines.bammetrics.BammetricsReport.mappingQualityPlotTables
Peter van 't Hof's avatar
Peter van 't Hof committed
27
import nl.lumc.sasc.biopet.utils.ConfigUtils
28
import nl.lumc.sasc.biopet.utils.rscript.{LinePlot, StackedBarPlot}
Peter van 't Hof's avatar
Peter van 't Hof committed
29
import nl.lumc.sasc.biopet.utils.summary.db.SummaryDb
Peter van 't Hof's avatar
Peter van 't Hof committed
30 31
import nl.lumc.sasc.biopet.utils.summary.db.SummaryDb.Implicts._
import nl.lumc.sasc.biopet.utils.summary.db.SummaryDb._
32
import nl.lumc.sasc.biopet.utils.summary.db.Schema._
33

34
import scala.collection.mutable.ArrayBuffer
35
import scala.concurrent.{Await, Future}
Peter van 't Hof's avatar
Peter van 't Hof committed
36
import scala.concurrent.duration.Duration
37

Peter van 't Hof's avatar
Peter van 't Hof committed
38
class BammetricsReport(val parent: Configurable) extends ReportBuilderExtension {
39
  def builder = BammetricsReport
40
}
41

42
/**
43 44 45 46
  * Object to create a report for [[BamMetrics]]
  *
  * Created by pjvan_thof on 3/30/15.
  */
47
object BammetricsReport extends ReportBuilder {
48

49
  /** Name of report */
50 51
  val reportName = "Bam Metrics"

52 53
  def pipelineName = "bammetrics"

54
  /** Root page for single BamMetrcis report */
Peter van 't Hof's avatar
Peter van 't Hof committed
55 56
  def indexPage: Future[ReportPage] =
    bamMetricsPage(summary, sampleId, libId).map { bamMetricsPage =>
57 58 59 60 61 62 63 64 65 66 67 68 69
      ReportPage(
        bamMetricsPage.subPages ::: List(
          "Versions" -> Future(
            ReportPage(List(),
                       List("Executables" -> ReportSection(
                         "/nl/lumc/sasc/biopet/core/report/executables.ssp")),
                       Map())),
          "Files" -> filesPage(sampleId, libId)
        ),
        List(
          "Report" -> ReportSection(
            "/nl/lumc/sasc/biopet/pipelines/bammetrics/bamMetricsFront.ssp")
        ) ::: bamMetricsPage.sections,
Peter van 't Hof's avatar
Peter van 't Hof committed
70 71 72
        Map()
      )
    }
73

74
  /** Generates a page with alignment stats */
Peter van 't Hof's avatar
Peter van 't Hof committed
75 76 77
  def bamMetricsPage(summary: SummaryDb,
                     sampleId: Option[Int],
                     libId: Option[Int],
Peter van 't Hof's avatar
Peter van 't Hof committed
78
                     metricsTag: String = "bammetrics"): Future[ReportPage] = Future {
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
    val wgsExecuted = summary.getStatsSize(runId,
                                           metricsTag,
                                           "wgs",
                                           sample = sampleId.map(SampleId),
                                           library = libId.map(LibraryId)) >= 1
    val rnaExecuted = summary.getStatsSize(runId,
                                           metricsTag,
                                           "rna",
                                           sample = sampleId.map(SampleId),
                                           library = libId.map(LibraryId)) >= 1

    val insertsizeMetrics = summary
      .getStatKeys(
        runId,
        metricsTag,
        "CollectInsertSizeMetrics",
        sampleId.map(SampleId).getOrElse(NoSample),
        libId.map(LibraryId).getOrElse(NoLibrary),
        Map("metrics" -> List("metrics"))
      )
Peter van 't Hof's avatar
Peter van 't Hof committed
99
      .exists(_._2.isDefined)
100

101 102 103 104 105 106 107 108
    val targetSettings = summary.getSettingKeys(
      runId,
      metricsTag,
      NoModule,
      sample = sampleId.map(SampleId).getOrElse(NoSample),
      library = libId.map(LibraryId).getOrElse(NoLibrary),
      Map("amplicon_name" -> List("amplicon_name"), "roi_name" -> List("roi_name"))
    )
109
    val targets = (
Peter van 't Hof's avatar
Peter van 't Hof committed
110 111
      targetSettings("amplicon_name"),
      targetSettings("roi_name")
Peter van 't Hof's avatar
Peter van 't Hof committed
112
    ) match {
113 114 115 116
      case (Some(amplicon: String), Some(roi: List[_])) => amplicon :: roi.map(_.toString)
      case (_, Some(roi: List[_])) => roi.map(_.toString)
      case _ => Nil
    }
117 118

    ReportPage(
Peter van 't Hof's avatar
Peter van 't Hof committed
119
      if (targets.isEmpty) List()
120 121 122 123 124 125 126 127 128
      else
        List(
          "Targets" -> Future.successful(
            ReportPage(
              List(),
              targets.map(t =>
                t -> ReportSection("/nl/lumc/sasc/biopet/pipelines/bammetrics/covstatsPlot.ssp",
                                   Map("target" -> Some(t)))),
              Map()))),
129
      List(
130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154
        "Summary" -> ReportSection(
          "/nl/lumc/sasc/biopet/pipelines/bammetrics/alignmentSummary.ssp"),
        "Mapping Quality" -> ReportSection(
          "/nl/lumc/sasc/biopet/pipelines/bammetrics/mappingQuality.ssp",
          Map("showPlot" -> true)),
        "Clipping" -> ReportSection("/nl/lumc/sasc/biopet/pipelines/bammetrics/clipping.ssp",
                                    Map("showPlot" -> true))
      ) ++
        (if (insertsizeMetrics)
           List(
             "Insert Size" -> ReportSection(
               "/nl/lumc/sasc/biopet/pipelines/bammetrics/insertSize.ssp",
               Map("showPlot" -> true)))
         else Nil) ++ (if (wgsExecuted)
                         List(
                           "Whole genome coverage" -> ReportSection(
                             "/nl/lumc/sasc/biopet/pipelines/bammetrics/wgsHistogram.ssp",
                             Map("showPlot" -> true)))
                       else Nil) ++
        (if (rnaExecuted)
           List(
             "Rna coverage" -> ReportSection(
               "/nl/lumc/sasc/biopet/pipelines/bammetrics/rnaHistogram.ssp",
               Map("showPlot" -> true)))
         else Nil),
155
      Map("metricsTag" -> metricsTag)
156 157 158
    )
  }

159
  /**
160
    * Generates the lines for alignmentSummaryPlot
161 162 163
    *
    * @param summary Summary class
    * @param sampleId Default it selects all sampples, when sample is giving it limits to selected sample
164
    *                     * @param libraryLevel Default false, when set true plot will be based on library stats instead of sample stats
165
    */
166
  def alignmentSummaryPlotLines(summary: SummaryDb,
167
                                sampleId: Option[Int] = None,
168
                                libraryLevel: Boolean = false): Seq[String] = {
Peter van 't Hof's avatar
Peter van 't Hof committed
169 170 171 172 173 174 175 176
    val statsPaths = Map(
      "Mapped" -> List("flagstats", "Mapped"),
      "Duplicates" -> List("flagstats", "Duplicates"),
      "All" -> List("flagstats", "All"),
      "NotPrimaryAlignment" -> List("flagstats", "NotPrimaryAlignment")
    )

    val results: Map[(Int, Option[Int]), Map[String, Option[Any]]] = if (libraryLevel) {
177 178
      summary
        .getStatsForLibraries(runId,
179 180 181 182
                              "bammetrics",
                              "bamstats",
                              sampleId = sampleId,
                              keyValues = statsPaths)
183 184 185 186
        .map(x => (x._1._1, Some(x._1._2)) -> x._2)
    } else
      summary
        .getStatsForSamples(runId,
187 188 189 190
                            "bammetrics",
                            "bamstats",
                            sample = sampleId.map(SampleId),
                            keyValues = statsPaths)
191
        .map(x => (x._1, None) -> x._2)
192
    val summaryPlotLines = ArrayBuffer[String]()
Peter van 't Hof's avatar
Peter van 't Hof committed
193

Peter van 't Hof's avatar
Peter van 't Hof committed
194
    for (((s, l), result) <- results) {
Peter van 't Hof's avatar
Peter van 't Hof committed
195
      val sampleName: String = summary.getSampleName(s).map(_.get)
196 197
      val libName: Option[String] =
        l.flatMap(x => Await.result(summary.getLibraryName(x), Duration.Inf))
198
      val sb = new StringBuffer()
199 200
      if (libName.isDefined) sb.append(sampleName + "-" + libName.get + "\t")
      else sb.append(sampleName + "\t")
Peter van 't Hof's avatar
Peter van 't Hof committed
201 202 203 204
      val mapped = ConfigUtils.any2long(result("Mapped"))
      val duplicates = ConfigUtils.any2long(result("Duplicates"))
      val total = ConfigUtils.any2long(result("All"))
      val secondary = ConfigUtils.any2long(result("NotPrimaryAlignment"))
Peter van 't Hof's avatar
Peter van 't Hof committed
205
      sb.append((mapped - duplicates - secondary) + "\t")
206
      sb.append(duplicates + "\t")
Peter van 't Hof's avatar
Peter van 't Hof committed
207
      sb.append((total - mapped) + "\t")
208
      sb.append(secondary)
209
      summaryPlotLines += sb.toString
210
    }
211 212 213 214 215 216 217 218 219 220 221 222 223 224
    summaryPlotLines
  }

  /**
    * Generate a stackbar plot for alignment stats
    *
    * @param outputDir OutputDir for the tsv and png file
    * @param prefix Prefix of the tsv and png file
    * @param summaryPlotLines A sequence of strings written to the summary tsv
    * @param libraryLevel Default false, when set true plot will be based on library stats instead of sample stats
    */
  def alignmentSummaryPlot(outputDir: File,
                           prefix: String,
                           summaryPlotLines: Seq[String],
225
                           libraryLevel: Boolean = false): Unit = {
226 227 228 229 230
    val tsvFile = new File(outputDir, prefix + ".tsv")
    val pngFile = new File(outputDir, prefix + ".png")
    val tsvWriter = new PrintWriter(tsvFile)
    if (libraryLevel) tsvWriter.print("Library") else tsvWriter.print("Sample")
    tsvWriter.println("\tMapped\tDuplicates\tUnmapped\tSecondary")
231

232 233 234
    for (line <- summaryPlotLines) {
      tsvWriter.println(line)
    }
235 236 237 238 239 240
    tsvWriter.close()

    val plot = new StackedBarPlot(null)
    plot.input = tsvFile
    plot.output = pngFile
    plot.ylabel = Some("Reads")
241
    plot.width = Some(200 + (summaryPlotLines.size * 10))
Peter van 't Hof's avatar
WIP  
Peter van 't Hof committed
242
    plot.title = Some("Aligned_reads")
243 244
    plot.runLocal()
  }
245

246
  /**
Ruben Vorderman's avatar
Ruben Vorderman committed
247
    * This is a generic method to create a Map that can be turned into a plot
248 249 250 251 252 253 254 255 256
    * @param libraryLevel If enabled the plots will show data per library
    * @param sampleId If set only this sample is shown
    * @param libraryId If set onlt this library is shown
    * @param statsPaths Paths in summary where the tables can be found
    * @param yKeyList Keys to search from, first has prio over second one
    * @param xKeyList Keys to search from, first has prio over second one
    * @param pipeline Query for the pipeline
    * @param module Query for the module
    */
257 258 259 260 261 262 263
  def summaryForPlot(summary: SummaryDb,
                     statsPaths: Map[String, List[String]],
                     yKeyList: List[String],
                     xKeyList: List[String],
                     pipeline: PipelineQuery,
                     module: ModuleQuery,
                     libraryLevel: Boolean = false,
264
                     sampleId: Option[Int] = None,
265
                     libraryId: Option[Int] = None): Array[Map[String, Array[Any]]] = {
Peter van 't Hof's avatar
Peter van 't Hof committed
266
    val results: Map[(Int, Option[Int]), Map[String, Option[Array[Any]]]] = if (libraryLevel) {
267 268 269 270 271 272 273 274
      summary
        .getStatsForLibraries(runId, pipeline, module, sampleId = sampleId, keyValues = statsPaths)
        .map(x =>
          (x._1._1, Some(x._1._2)) -> x._2.map(x =>
            x._1 -> x._2.map(ConfigUtils.any2list(_).toArray)))
    } else
      summary
        .getStatsForSamples(runId,
275 276 277 278
                            pipeline,
                            module,
                            sample = sampleId.map(SampleId),
                            keyValues = statsPaths)
279
        .map(x => (x._1, None) -> x._2.map(x => x._1 -> x._2.map(ConfigUtils.any2list(_).toArray)))
Peter van 't Hof's avatar
Peter van 't Hof committed
280 281
    val tables: Array[Map[String, Array[Any]]] = results.map {
      case ((sample, library), map) =>
282 283
        val sampleName = Await
          .result(summary.getSampleName(sample), Duration.Inf)
Peter van 't Hof's avatar
Peter van 't Hof committed
284
          .getOrElse(throw new IllegalStateException("Sample must be there"))
285 286
        val libraryName =
          library.flatMap(l => Await.result(summary.getLibraryName(l), Duration.Inf))
Peter van 't Hof's avatar
Peter van 't Hof committed
287 288
        val yKey = yKeyList.find(x => map.contains(x) && map(x).isDefined).getOrElse("none")
        val xKey = xKeyList.find(x => map.contains(x) && map(x).isDefined).getOrElse("none")
Peter van 't Hof's avatar
Peter van 't Hof committed
289
        Map(
Peter van 't Hof's avatar
Peter van 't Hof committed
290 291 292 293
          yKeyList.head -> map.getOrElse(yKey, None).getOrElse(Array()),
          (sampleName + libraryName.map("-" + _).getOrElse("")) -> map
            .getOrElse(xKey, None)
            .getOrElse(Array())
Peter van 't Hof's avatar
Peter van 't Hof committed
294
        )
Peter van 't Hof's avatar
Peter van 't Hof committed
295
    }.toArray
296 297 298 299 300 301 302
    tables
  }

  /**
    * This is a generic method to create plots
    * @param outputDir Outputdir of the plot
    * @param prefix Files will start with this name
Ruben Vorderman's avatar
Ruben Vorderman committed
303
    * @param plotTables Tables to be written
304 305 306 307 308 309 310 311 312
    * @param yKeyList Keys to search from, first has prio over second one
    * @param xKeyList Keys to search from, first has prio over second one
    * @param xlabel X label shown on the plot
    * @param ylabel Y label shown on the plot
    * @param title Title of the plot
    * @param removeZero
    */
  def writePlotFromSummary(outputDir: File,
                           prefix: String,
Ruben Vorderman's avatar
Ruben Vorderman committed
313
                           plotTables: Array[Map[String, Array[Any]]],
314
                           yKeyList: List[String],
315
                           xKeyList: List[String],
316 317 318 319 320 321
                           xlabel: Option[String] = None,
                           ylabel: Option[String] = None,
                           title: Option[String] = None,
                           removeZero: Boolean = true): Unit = {
    val tsvFile = new File(outputDir, prefix + ".tsv")
    val pngFile = new File(outputDir, prefix + ".png")
Peter van 't Hof's avatar
Peter van 't Hof committed
322

Ruben Vorderman's avatar
Ruben Vorderman committed
323
    writeTableToTsv(tsvFile, mergeTables(plotTables, yKeyList.head), yKeyList.head)
Peter van 't Hof's avatar
Peter van 't Hof committed
324

325 326 327 328 329 330
    LinePlot(
      tsvFile,
      pngFile,
      xlabel = xlabel,
      ylabel = ylabel,
      title = title,
Ruben Vorderman's avatar
Ruben Vorderman committed
331
      hideLegend = plotTables.size > 40,
332 333 334
      /* changed from results.size. Original results in summaryForPlot*/
      removeZero = removeZero
    ).runLocal()
Peter van 't Hof's avatar
Peter van 't Hof committed
335 336
  }

337 338 339
  def insertSizePlotTables(summary: SummaryDb,
                           libraryLevel: Boolean = false,
                           sampleId: Option[Int] = None,
Ruben Vorderman's avatar
Ruben Vorderman committed
340 341 342 343 344
                           libraryId: Option[Int] = None): Array[Map[String, Array[Any]]] = {
    val statsPaths = Map(
      "insert_size" -> List("histogram", "insert_size"),
      "count" -> List("histogram", "All_Reads.fr_count")
    )
Ruben Vorderman's avatar
Ruben Vorderman committed
345 346 347 348 349 350 351 352 353
    summaryForPlot(summary,
                   statsPaths,
                   "insert_size" :: Nil,
                   "count" :: Nil,
                   "bammetrics",
                   "CollectInsertSizeMetrics",
                   libraryLevel,
                   sampleId,
                   libraryId)
Ruben Vorderman's avatar
Ruben Vorderman committed
354
  }
355

356
  /**
357 358 359 360
    * Generate a line plot for insertsize
    *
    * @param outputDir OutputDir for the tsv and png file
    * @param prefix Prefix of the tsv and png file
Ruben Vorderman's avatar
Ruben Vorderman committed
361
    * @param insertSizePlotTables Plot map generated by insertSizePlotTables function.
362
    */
Peter van 't Hof's avatar
Peter van 't Hof committed
363
  def insertSizePlot(outputDir: File,
Peter van 't Hof's avatar
Peter van 't Hof committed
364
                     prefix: String,
365
                     insertSizePlotTables: Array[Map[String, Array[Any]]]): Unit = {
Ruben Vorderman's avatar
Ruben Vorderman committed
366 367 368 369 370 371 372 373
    writePlotFromSummary(outputDir,
                         prefix,
                         insertSizePlotTables,
                         "insert_size" :: Nil,
                         "count" :: Nil,
                         "Insert size",
                         "Reads",
                         "Insert size")
Peter van 't Hof's avatar
Peter van 't Hof committed
374
  }
Peter van 't Hof's avatar
Peter van 't Hof committed
375

376 377 378 379
  def mappingQualityPlotTables(summary: SummaryDb,
                               libraryLevel: Boolean = false,
                               sampleId: Option[Int] = None,
                               libraryId: Option[Int] = None): Array[Map[String, Array[Any]]] = {
Peter van 't Hof's avatar
Peter van 't Hof committed
380 381 382
    val statsPaths = Map(
      "mapping_quality" -> List("mapping_quality", "histogram", "values"),
      "count" -> List("mapping_quality", "histogram", "counts")
383
    )
Ruben Vorderman's avatar
Ruben Vorderman committed
384 385 386 387 388 389 390 391 392
    summaryForPlot(summary,
                   statsPaths,
                   "mapping_quality" :: Nil,
                   "count" :: Nil,
                   "bammetrics",
                   "bamstats",
                   libraryLevel,
                   sampleId,
                   libraryId)
393 394
  }

395 396 397 398 399 400 401 402 403 404 405 406
  def mappingQualityPlot(outputDir: File,
                         prefix: String,
                         mappingQualityPlotTables: Array[Map[String, Array[Any]]]): Unit = {
    writePlotFromSummary(outputDir,
                         prefix,
                         mappingQualityPlotTables,
                         "mapping_quality" :: Nil,
                         "count" :: Nil,
                         "Mapping Quality",
                         "Reads",
                         "Mapping Quality")
  }
Ruben Vorderman's avatar
Ruben Vorderman committed
407

408 409 410
  def clippingPlotTables(summary: SummaryDb,
                         libraryLevel: Boolean = false,
                         sampleId: Option[Int] = None,
Ruben Vorderman's avatar
Ruben Vorderman committed
411
                         libraryId: Option[Int] = None): Array[Map[String, Array[Any]]] = {
Peter van 't Hof's avatar
Peter van 't Hof committed
412 413 414
    val statsPaths = Map(
      "clipping" -> List("clipping", "histogram", "values"),
      "count" -> List("clipping", "histogram", "counts")
415
    )
Ruben Vorderman's avatar
Ruben Vorderman committed
416 417 418 419 420 421 422 423 424
    summaryForPlot(summary,
                   statsPaths,
                   "clipping" :: Nil,
                   "count" :: Nil,
                   "bammetrics",
                   "bamstats",
                   libraryLevel,
                   sampleId,
                   libraryId)
425

Ruben Vorderman's avatar
Ruben Vorderman committed
426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447
  }
  def clippingPlot(outputDir: File,
                   prefix: String,
                   clippingPlotTables: Array[Map[String, Array[Any]]]): Unit = {
    writePlotFromSummary(outputDir,
                         prefix,
                         clippingPlotTables,
                         "clipping" :: Nil,
                         "count" :: Nil,
                         "Clipping",
                         "Reads",
                         "Clipping")
  }

  def wgsHistogramPlotTables(summary: SummaryDb,
                             libraryLevel: Boolean = false,
                             sampleId: Option[Int] = None,
                             libraryId: Option[Int] = None): Array[Map[String, Array[Any]]] = {
    val statsPaths = Map(
      "coverage" -> List("histogram", "coverage"),
      "count" -> List("histogram", "count"),
      "high_quality_coverage_count" -> List("histogram", "high_quality_coverage_count")
448
    )
Ruben Vorderman's avatar
Ruben Vorderman committed
449 450 451 452 453 454 455 456 457
    summaryForPlot(summary,
                   statsPaths,
                   "coverage" :: Nil,
                   "count" :: "high_quality_coverage_count" :: Nil,
                   "bammetrics",
                   "wgs",
                   libraryLevel,
                   sampleId,
                   libraryId)
458 459
  }

460
  /**
461 462 463 464
    * Generate a line plot for wgs coverage
    *
    * @param outputDir OutputDir for the tsv and png file
    * @param prefix Prefix of the tsv and png file
Ruben Vorderman's avatar
Ruben Vorderman committed
465
    * @param wgsHistogramPlotTables Plot map generated by wgsHistogramPlotTables function.
466
    */
Peter van 't Hof's avatar
Peter van 't Hof committed
467
  def wgsHistogramPlot(outputDir: File,
Peter van 't Hof's avatar
Peter van 't Hof committed
468
                       prefix: String,
Ruben Vorderman's avatar
Ruben Vorderman committed
469 470 471 472 473 474 475 476 477
                       wgsHistogramPlotTables: Array[Map[String, Array[Any]]]): Unit = {
    writePlotFromSummary(outputDir,
                         prefix,
                         wgsHistogramPlotTables,
                         "coverage" :: Nil,
                         "count" :: "high_quality_coverage_count" :: Nil,
                         "Coverage",
                         "Bases",
                         "Whole genome coverage")
Peter van 't Hof's avatar
Peter van 't Hof committed
478
  }
Peter van 't Hof's avatar
Peter van 't Hof committed
479

Ruben Vorderman's avatar
Ruben Vorderman committed
480 481 482 483 484 485 486 487
  def rnaHistogramPlotTables(summary: SummaryDb,
                             libraryLevel: Boolean = false,
                             sampleId: Option[Int] = None,
                             libraryId: Option[Int] = None): Array[Map[String, Array[Any]]] = {
    val statsPaths = Map(
      "position" -> List("histogram", "normalized_position"),
      "count" -> List("histogram", "All_Reads.normalized_coverage")
    )
488 489 490 491 492 493 494 495 496
    summaryForPlot(summary,
                   statsPaths,
                   "position" :: Nil,
                   "count" :: Nil,
                   "bammetrics",
                   "rna",
                   libraryLevel,
                   sampleId,
                   libraryId)
Ruben Vorderman's avatar
Ruben Vorderman committed
497
  }
498

Peter van 't Hof's avatar
Peter van 't Hof committed
499
  /**
500 501 502 503
    * Generate a line plot for rna coverage
    *
    * @param outputDir OutputDir for the tsv and png file
    * @param prefix Prefix of the tsv and png file
Ruben Vorderman's avatar
Ruben Vorderman committed
504
    * @param rnaHistogramPlotTables Plot map generated by rnaHistogramPlotTables function.
505
    */
Peter van 't Hof's avatar
Peter van 't Hof committed
506 507
  def rnaHistogramPlot(outputDir: File,
                       prefix: String,
Ruben Vorderman's avatar
Ruben Vorderman committed
508
                       rnaHistogramPlotTables: Array[Map[String, Array[Any]]],
Peter van 't Hof's avatar
Peter van 't Hof committed
509
                       libraryLevel: Boolean = false,
Peter van 't Hof's avatar
Peter van 't Hof committed
510 511
                       sampleId: Option[Int] = None,
                       libraryId: Option[Int] = None): Unit = {
Ruben Vorderman's avatar
Ruben Vorderman committed
512
    writePlotFromSummary(outputDir,
513 514 515 516 517 518 519
                         prefix,
                         rnaHistogramPlotTables,
                         "position" :: Nil,
                         "count" :: Nil,
                         "Relative position",
                         "Coverage",
                         "RNA coverage")
Peter van 't Hof's avatar
Peter van 't Hof committed
520 521
  }

Peter van 't Hof's avatar
Peter van 't Hof committed
522
  def mergeTables(tables: Array[Map[String, Array[Any]]],
523 524
                  mergeColumn: String,
                  defaultValue: Any = 0): Map[String, Array[Any]] = {
Peter van 't Hof's avatar
Peter van 't Hof committed
525 526
    val keys = tables.flatMap(x => x(mergeColumn)).distinct
    (for (table <- tables; (columnKey, columnValues) <- table if columnKey != mergeColumn) yield {
527 528
      columnKey -> keys.map(x =>
        table(mergeColumn).zip(columnValues).toMap.getOrElse(x, defaultValue))
Peter van 't Hof's avatar
Peter van 't Hof committed
529 530 531 532
    }).toMap + (mergeColumn -> keys)
  }

  def writeTableToTsv(tsvFile: File, table: Map[String, Array[Any]], firstColumn: String): Unit = {
533
    require(table.map(_._2.length).toList.distinct.size == 1,
534
            "Not all values has the same number or rows")
Peter van 't Hof's avatar
Peter van 't Hof committed
535 536 537
    val keys = table.keys.filterNot(_ == firstColumn).toList.sorted
    val writer = new PrintWriter(tsvFile)
    writer.println((firstColumn :: keys).mkString("\t"))
Peter van 't Hof's avatar
Peter van 't Hof committed
538 539 540
    table(firstColumn).zipWithIndex.foreach {
      case (c, i) =>
        writer.println((c :: keys.map(x => table(x)(i))).mkString("\t"))
Peter van 't Hof's avatar
Peter van 't Hof committed
541 542 543
    }
    writer.close()
  }
544
}