Merge branch 'issue3087-annotation_fix' into 'dev'

Issue3087 annotation fix

Closes #87

See merge request BICF/Astrocyte/chipseq_analysis!72

workflow/scripts/annotate_peaks.R

@@ -43,13 +43,14 @@ names(files) <- design$Condition
 peaks <- lapply(files, readPeakFile, as = "GRanges", header = FALSE)
 peakAnnoList <- lapply(peaks, annotatePeak, TxDb=txdb, tssRegion=c(-3000, 3000), verbose=FALSE)
 
-column_names <- c("chr", "start", "end", "width", "strand_1", "name", "score", "strand", "signalValue",
+column_names <- c("geneId","chr", "start", "end", "width", "strand_1", "name", "score", "strand", "signalValue",
                   "pValue", "qValue", "peak", "annotation", "geneChr", "geneStart", "geneEnd",
-                  "geneLength" ,"geneStrand", "geneId", "transcriptId", "distanceToTSS", "symbol")
+                  "geneLength" ,"geneStrand", "transcriptId", "distanceToTSS", "symbol")
 
 for(index in c(1:length(peakAnnoList))) {
   filename <- paste(names(peaks)[index], ".chipseeker_annotation.tsv", sep="")
   df <- as.data.frame(peakAnnoList[[index]])
+  df$geneId <- sapply(strsplit(as.character(df$geneId), split = "\\."), "[[", 1)
   df_final <- merge(df, sym, by.x="geneId", by.y="ensembl", all.x=T)
   colnames(df_final) <- column_names
   write.table(df_final[ , !(names(df_final) %in% c('strand_1'))], filename, sep="\t" ,quote=F, row.names=F)

workflow/tests/test_annotate_peaks.py

@@ -25,6 +25,10 @@ def test_annotation_singleend():
     annotation_file = test_output_path + 'ENCSR238SGC.chipseeker_annotation.tsv'
     assert os.path.exists(annotation_file)
     assert utils.count_lines(annotation_file) >= 149284
+    df = pd.read_csv(annotation_file, sep = "\t", header = 0)
+    print(df.head())
+    #assert df['symbol'].notna().all()
+    assert not(df['symbol'].isnull().values.any())
 
 
 @pytest.mark.pairedend
@@ -42,3 +46,7 @@ def test_annotation_pairedend():
     annotation_file = test_output_path + 'ENCSR729LGA.chipseeker_annotation.tsv'
     assert os.path.exists(annotation_file)
     assert utils.count_lines(annotation_file) >= 25367
+    df = pd.read_csv(annotation_file, sep = "\t", header = 0)
+    print(df.head())
+    #assert df['symbol'].notna().all()
+    assert not(df['symbol'].isnull().values.any())