6d4fc5c83c417afe88dc436f03117f0c128dfda9
kent
  Tue Jan 19 09:40:18 2021 -0800
Updating mapping.bed and statistics

diff --git src/hg/makeDb/doc/hg19.txt src/hg/makeDb/doc/hg19.txt
index afc6371..a28e1de 100644
--- src/hg/makeDb/doc/hg19.txt
+++ src/hg/makeDb/doc/hg19.txt
@@ -34245,31 +34245,31 @@
     ln -s /hive/data/genomes/hg19/bed/singleCell/skinSoleBoldo/bbi/$s.stats /gbdb/hg19/bbi/skinSoleBoldo/
 end
 
 # Add the bits from clust/*.ra and clust/*.colors to hg19/trackDb.ra and you should be good.
 rm -f tracks.ra
 foreach s (cell_type donor age age_cell_type)
     grep -v barChartColors clust/$s.ra >>tracks.ra
     cat clust/$s.colors >> tracks.ra
     echo transformFunc NONE >> tracks.ra
     echo barChartLimit 2 >> tracks.ra
     echo "" >> tracks.ra
 end
 
 
 #############################################################################
-# fetalGeneAtlas JimK 01-14-2020
+# fetalGeneAtlas JimK 01-19-2020
 ############################################################################
 # This is the RNA-seq part of the data set described in 
 # "A human cell atlas of fetal gene expression" by Cao, Day et al
 # Science 13 Nove 2020.   This was imported from Cell Browser
 
 # Create directory for work.
 
 mkdir -p /hive/data/genomes/hg19/bed/singleCell/fetalGeneAtlas
 cd /hive/data/genomes/hg19/bed/singleCell/fetalGeneAtlas
 
 # Create output dir for binaries
 mkdir bbi 
 
 # link in in files from cell browser
 ln -s /hive/data/inside/cells/datasets/fetal-gene-atlas/genes/all/meta.tsv .
@@ -34287,30 +34287,32 @@
     Main_cluster_name clust/cell_type.matrix bbi/cell_type.stats \
     Assay clust/Assay.matrix bbi/Assay.stats \
     Experiment_batch clust/Experiment_batch.matrix bbi/Experiment_batch.stats \
     Fetus_id clust/donor.matrix bbi/donor.stats \
     Organ clust/Organ.matrix bbi/Organ.stats \
     Organ_cell_lineage clust/Organ_cell_lineage.matrix bbi/Organ_cell_lineage.stats \
     RT_group clust/RT_group.matrix bbi/RT_group.stats \
     sex clust/sex.matrix bbi/sex.stats 
 
 # Get the first column (the genes) out of expression matrix.  
 cut -f 1 clust/cell_type.matrix > gene.lst
 
 
 # Figure out the geneset they used and generate mapping file
 gencodeVersionForGenes gene.lst /hive/data/inside/geneSymVerTx.tsv -bed=mapping.bed
+# best is gencodeV19 as id on hg19 with 60284 of 63562 (94.8428%) hits
+
 
 # Turn some into barChart, and then bigBarChart
 foreach s (cell_type Assay Experiment_batch donor Organ Organ_cell_lineage RT_group sex)
     matrixToBarChartBed clust/$s.matrix mapping.bed clust/$s.bed -stats=bbi/$s.stats -trackDb=clust/$s.ra
     bedSort clust/$s.bed clust/$s.bed
     bedToBigBed clust/$s.bed /hive/data/genomes/hg19/chrom.sizes bbi/$s.bb -type=bed6+3 -as=/cluster/home/kent/src/hg/lib/simpleBarChartBed.as
 end
 
 # Make up special colors for cell_type.  First manually create two column 
 # file that relates at least some of sample labels to cell types we have colors for.
 # Call this file clust/cell_type.labels.  
 matrixClusterColumns clust/cell_type.matrix clust/cell_type.labels cluster clust/cell_type.unnormed clust/cell_type.restats
 matrixNormalize column sum clust/cell_type.unnormed clust/cell_type.ref
 hcaColorCells clust/cell_type.ref ../typeColors.tsv clust/cell_type.matrix clust/cell_type.refStats -trackDb=clust/cell_type.colors -stats=bbi/cell_type.stats