Merge discvr-22.3 to develop

singlecell/api-src/org/labkey/api/singlecell/pipeline/AbstractSingleCellPipelineStep.java

@@ -348,38 +348,52 @@ public static void executeR(SequenceOutputHandler.JobContext ctx, String dockerC
  localBashScript.delete();
  }
 
- protected String prepareValueForR(SeuratToolParameter pd)
+ protected void addParameterVariables(SeuratToolParameter pd, List<String> body)
  {
  String val = StringUtils.trimToNull(pd.extractValue(getPipelineCtx().getJob(), getProvider(), getStepIdx()));
  if (val == null)
  {
-return "NULL";
+body.add((pd.getVariableName() + " <- NULL"));
  }
  else if ("false".equals(val))
  {
-return "FALSE";
+body.add((pd.getVariableName() + " <- FALSE"));
  }
  else if ("true".equals(val))
  {
-return "TRUE";
+body.add((pd.getVariableName() + " <- TRUE"));
  }
  else if (NumberUtils.isCreatable(val))
  {
-return val;
+body.add((pd.getVariableName() + " <- " + val));
  }
  else if ("sequenceanalysis-trimmingtextarea".equals(pd.getFieldXtype()))
  {
  val = val.replace("'", "\\\'");
- String[] vals = val.split(pd.getDelimiter());
- return "c('" + StringUtils.join(vals, "','") + "')";
+ serializeMultiValueParam(pd, body, val);
  }
  else if (pd.isMultiValue())
  {
- String[] vals = val.split(pd.getDelimiter());
- return "c('" + StringUtils.join(vals, "','") + "')";
+ serializeMultiValueParam(pd, body, val);
  }
+ else
+ {
+ body.add((pd.getVariableName() + " <- '" + val + "'"));
+ }
+ }
 
- return "'" + val + "'";
+ private void serializeMultiValueParam(SeuratToolParameter pd, List<String> body, String val)
+ {
+ String[] vals = val.split(pd.getDelimiter());
+ final int batchSize = 75;
+ int numBatches = (int)Math.ceil((double)vals.length / batchSize);
+
+ for (int i=0;i<numBatches;i++)
+ {
+ int start = (i * batchSize);
+ int end = Math.min(start + batchSize, vals.length);
+ body.add(pd.getVariableName() + " <- " + "c(" + (i == 0 ? "" : pd.getVariableName() + ", ") + "'" + StringUtils.join(Arrays.copyOfRange(vals, start, end), "','") + "')");
+ }
  }
 
  protected List<String> loadChunkFromFile() throws PipelineJobException
@@ -423,7 +437,7 @@ protected Chunk createParamChunk(SequenceOutputHandler.JobContext ctx, List<Seur
  SeuratToolParameter stp = (SeuratToolParameter)pd;
  if (stp.shouldIncludeInMarkdown(getPipelineCtx().getJob(), getProvider(), getStepIdx()))
  {
-body.add(((SeuratToolParameter) pd).getVariableName() + " <- " + prepareValueForR(stp));
+addParameterVariables(stp, body);
  }
  }
  }

singlecell/api-src/org/labkey/api/singlecell/pipeline/SeuratToolParameter.java

@@ -50,7 +50,7 @@ public String getVariableName()
  return _rName == null ? getName() : _rName;
  }
 
- public boolean shouldIncludeInMarkdown(PipelineJob job, PipelineStepProvider provider, int stepIdx)
+ public boolean shouldIncludeInMarkdown(PipelineJob job, PipelineStepProvider<?> provider, int stepIdx)
  {
  if (!_includeIfEmptyOrNull)
  {

singlecell/resources/chunks/TcrFilter.R

@@ -18,7 +18,11 @@ for (datasetId in names(seuratObjects)) {
 
  cdr3ForLocus <- gsub(cdr3ForLocus, pattern = paste0(fieldName, ':'), replacement = '')
  matchingCells <- sapply(seuratObj@meta.data[[fieldName]], function(x){
- values <- unlist(strsplit(x, split = ','))
+ if (is.na(x)) {
+ return(FALSE)
+ }
+
+ values <- unlist(strsplit(as.character(x), split = ','))
  return(length(intersect(values, cdr3ForLocus)) != 0)
  })
 
@@ -37,8 +41,9 @@ for (datasetId in names(seuratObjects)) {
  if (all(is.null(cellsToKeep))) {
  print('There were no matching cells')
  } else {
+ print(paste0('Total passing cells: ', length(cellsToKeep)))
  seuratObj <- subset(seuratObj, cells = cellsToKeep)
-#saveData(seuratObj, datasetId)
+ saveData(seuratObj, datasetId)
  totalPassed <- totalPassed + 1
  }
 

singlecell/src/org/labkey/singlecell/SingleCellModule.java

@@ -188,7 +188,7 @@ public static void registerPipelineSteps()
  SequencePipelineService.get().registerPipelineStep(new CheckExpectations.Provider());
  SequencePipelineService.get().registerPipelineStep(new CommonFilters.Provider());
  SequencePipelineService.get().registerPipelineStep(new RunVision.Provider());
- SequencePipelineService.get().registerPipelineStep(new NimbleAppend.Provider());
+ SequencePipelineService.get().registerPipelineStep(new AppendNimble.Provider());
  SequencePipelineService.get().registerPipelineStep(new AppendTcr.Provider());
  SequencePipelineService.get().registerPipelineStep(new TcrFilter.Provider());
 

singlecell/src/org/labkey/singlecell/pipeline/singlecell/NimbleAppend.java → singlecell/src/org/labkey/singlecell/pipeline/singlecell/AppendNimble.java

@@ -15,9 +15,9 @@
 import java.util.List;
 import java.util.Set;
 
-public class NimbleAppend extends AbstractRDiscvrStep
+public class AppendNimble extends AbstractRDiscvrStep
 {
- public NimbleAppend(PipelineContext ctx, NimbleAppend.Provider provider)
+ public AppendNimble(PipelineContext ctx, AppendNimble.Provider provider)
  {
  super(provider, ctx);
  }
@@ -35,9 +35,9 @@ public Provider()
 
 
  @Override
- public NimbleAppend create(PipelineContext ctx)
+ public AppendNimble create(PipelineContext ctx)
  {
- return new NimbleAppend(ctx, this);
+ return new AppendNimble(ctx, this);
  }
  }
 
@@ -62,6 +62,7 @@ protected Chunk createParamChunk(SequenceOutputHandler.JobContext ctx, List<Seur
  ret.bodyLines.add("nimbleGenomes <- list(");
  String genomeStr = getProvider().getParameterByName("nimbleGenomes").extractValue(getPipelineCtx().getJob(), getProvider(), getStepIdx(), String.class);
  JSONArray json = new JSONArray(genomeStr);
+ String delim = "";
  for (int i = 0; i < json.length(); i++)
  {
  JSONArray arr = json.getJSONArray(i);
@@ -72,7 +73,8 @@ protected Chunk createParamChunk(SequenceOutputHandler.JobContext ctx, List<Seur
 
  int genomeId = arr.getInt(0);
  String targetAssay = arr.getString(1);
- ret.bodyLines.add("\t" + genomeId + " = " + targetAssay);
+ ret.bodyLines.add("\t" + delim + "'" + genomeId + "' = '" + targetAssay + "'");
+ delim = ",";
  }
  ret.bodyLines.add(")");
 

singlecell/src/org/labkey/singlecell/pipeline/singlecell/CheckExpectations.java

@@ -72,10 +72,15 @@ protected Chunk createParamChunk(SequenceOutputHandler.JobContext ctx, List<Seur
 
  for (SeuratObjectWrapper so : inputObjects)
  {
+ if (so.getSequenceOutputFile().getReadset() == null)
+ {
+ throw new PipelineJobException("Unable to find readset for outputfile: " + so.getSequenceOutputFileId() + ". This set requires single-dataset inputs. Removing this step may be a solution.");
+ }
+
  Readset parentReadset = ctx.getSequenceSupport().getCachedReadset(so.getSequenceOutputFile().getReadset());
  if (parentReadset == null)
  {
- throw new PipelineJobException("Unable to find readset for outputfile: " + so.getSequenceOutputFileId());
+ throw new PipelineJobException("Unable to find readset for outputfile: " + so.getSequenceOutputFileId() + ". This set requires single-dataset inputs. Removing this step may be a solution.");
  }
 
  Set<String> htosPerReadset = CellHashingServiceImpl.get().getHtosForParentReadset(parentReadset.getReadsetId(), ctx.getSourceDirectory(), ctx.getSequenceSupport(), false);