Pass GTF file to OOSAP/Seurat

Author: bbimber

SequenceAnalysis/src/org/labkey/sequenceanalysis/run/alignment/CellRangerWrapper.java

@@ -60,6 +60,8 @@
 
 public class CellRangerWrapper extends AbstractCommandWrapper
 {
+    public static final String GTF_FILE = "GTF File";
+
     public CellRangerWrapper(@Nullable Logger logger)
     {
         super(logger);
@@ -93,10 +95,7 @@ public Provider()
                         put("extensions", Arrays.asList("gtf"));
                         put("width", 400);
                         put("allowBlank", false);
-                    }}, null),
-                    ToolParameterDescriptor.create("premrna", "Use pre-mRNA GTF", "Normally, reads are only counted if they overlap exons.  If selected, the pipeline will convert the GTF to list all transcript intervals as exon, meaning reads within introns will be counted as well.  This could be useful for single-nuclei sequencing (which captures pre-mRNA), or if your GTF exon annotations may be lacking.", "checkbox", new JSONObject(){{
-
-                    }}, false)
+                    }}, null)
             ), PageFlowUtil.set("sequenceanalysis/field/GenomeFileSelectorField.js"), "https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/what-is-cell-ranger", true, false, ALIGNMENT_MODE.MERGE_THEN_ALIGN);
         }
 
@@ -137,7 +136,6 @@ public String getAlignmentDescription()
 
         protected static String getAlignDescription(PipelineStepProvider provider, PipelineContext ctx, int stepIdx, boolean addAligner)
         {
-            boolean isPreMrna = isPreMrna(provider, ctx, stepIdx);
             Integer gtfId = provider.getParameterByName("gtfFile").extractValue(ctx.getJob(), provider, stepIdx, Integer.class);
             File gtfFile = ctx.getSequenceSupport().getCachedData(gtfId);
             if (gtfFile == null)
@@ -160,11 +158,6 @@ protected static String getAlignDescription(PipelineStepProvider provider, Pipel
                 lines.add("GTF: " + gtfFile.getName());
             }
 
-            if (isPreMrna)
-            {
-                lines.add("Converted to pre-mRNA GTF");
-            }
-
             return lines.isEmpty() ? null : StringUtils.join(lines, '\n');
         }
 
@@ -177,14 +170,7 @@ public String getIndexCachedDirName(PipelineJob job)
                 throw new IllegalArgumentException("Missing gtfFile parameter");
             }
 
-            boolean premrna = isPreMrna(getProvider(), getPipelineCtx(), getStepIdx());
-
-            return "cellRanger-" + gtfId + (premrna ? "-premrna" : "");
-        }
-
-        private static boolean isPreMrna(PipelineStepProvider provider, PipelineContext ctx, int stepIdx)
-        {
-            return provider.getParameterByName("premrna").extractValue(ctx.getJob(), provider, stepIdx, Boolean.class, false);
+            return "cellRanger-" + gtfId;
         }
 
         @Override
@@ -226,13 +212,6 @@ public IndexOutput createIndex(ReferenceGenome referenceGenome, File outputDir)
 
                 File gtfEdit = new File(indexDir.getParentFile(), FileUtil.getBaseName(gtfFile) + ".geneId.gtf");
 
-                //See: https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/advanced/references
-                boolean premrna = getProvider().getParameterByName("premrna").extractValue(getPipelineCtx().getJob(), getProvider(), getStepIdx(), Boolean.class, false);
-                if (premrna)
-                {
-                    getPipelineCtx().getLogger().debug("Creating a pre-mRNA version of the GTF");
-                }
-
                 try (CSVReader reader = new CSVReader(Readers.getReader(gtfFile), '\t', CSVWriter.NO_QUOTE_CHARACTER); CSVWriter writer = new CSVWriter(PrintWriters.getPrintWriter(gtfEdit), '\t', CSVWriter.NO_QUOTE_CHARACTER, CSVWriter.NO_ESCAPE_CHARACTER))
                 {
                     String[] line;
@@ -251,12 +230,6 @@ public IndexOutput createIndex(ReferenceGenome referenceGenome, File outputDir)
                             continue;
                         }
 
-                        if (premrna && "transcript".equalsIgnoreCase(line[2]))
-                        {
-                            exonsAdded++;
-                            line[2] = "exon";
-                        }
-
                         writer.writeNext(line);
                     }
                 }
@@ -270,12 +243,7 @@ public IndexOutput createIndex(ReferenceGenome referenceGenome, File outputDir)
                     getPipelineCtx().getLogger().info("dropped " + linesDropped + " lines lacking gene_id, transcript_id, or with an empty value for gene_id/transcript_id");
                 }
 
-                if (premrna)
-                {
-                    getPipelineCtx().getLogger().info("total transcripts converted to exon: " + exonsAdded);
-                }
-
-                boolean useAlternateGtf = linesDropped > 0 || premrna;
+                boolean useAlternateGtf = linesDropped > 0;
                 if (useAlternateGtf)
                 {
                     gtfFile = gtfEdit;
@@ -364,6 +332,10 @@ public AlignmentOutput performAlignment(Readset rs, File inputFastq1, @Nullable
                 args.add("--sample=" + StringUtils.join(sampleNames, ","));
             }
 
+            Integer gtfId = getProvider().getParameterByName("gtfFile").extractValue(getPipelineCtx().getJob(), getProvider(), getStepIdx(), Integer.class);
+            File gtfFile = getPipelineCtx().getSequenceSupport().getCachedData(gtfId);
+            output.addInput(gtfFile, GTF_FILE);
+
             File indexDir = AlignerIndexUtil.getWebserverIndexDir(referenceGenome, getIndexCachedDirName(getPipelineCtx().getJob()));
             args.add("--transcriptome=" + indexDir.getPath());