7b3269bd1e75158ddf049cd109f7f4e1383da0ec
kate
  Tue May 14 10:36:23 2019 -0700
Color subtracks by DNase similarity. refs #23243

diff --git src/hg/makeDb/doc/encode3/tfbs.txt src/hg/makeDb/doc/encode3/tfbs.txt
index be450bf..f60f3ec 100644
--- src/hg/makeDb/doc/encode3/tfbs.txt
+++ src/hg/makeDb/doc/encode3/tfbs.txt
@@ -363,30 +363,54 @@
 
 # Fix up subGroup members w/ punctuation and initial numbers:
 #       Peyer's_patch -> Peyers_patch
 #       NT2/D1 -> NT2_D1
 #       22Rv1 -> X22Rv1
 #       MM.1S -> MM_1S
 
 # rename tables (encode3Tfbs -> encTfChipPk)
 
 hgsql hg19 -e 'show tables like "encode3TfbsPk%"' > tables.old.txt
 sed -e 's/^/alter table /' -e 's/$/ rename to /' tables.old.txt > rename.1.sql
 sed -e 's/encode3TfbsPk/encTfChipPk/' tables.old.txt | paste rename.1.sql -  | \
     sed 's/$/;/' > rename.sql
 hgsql hg19 < rename.sql
 
+# add colors to tier1-3 cell experiments (using ENCODE 2 color conventions)
+#
+# GM12878           color 153,38,0
+# H1-hESC           color 0,107,27
+# K562              color 46,0,184
+# HeLa-S3           color 0,119,158
+# HepG2             color 189,0,157
+# HUVEC             color 224,75,0
+
+raToLines encode3.ra encode3.lines
+sed -e '/GM12878/s/$/| color 153,38,0/' \
+    -e '/H1-hESC/s/$/| color 0,107,27/' \
+    -e '/K562/s/$/| color 46,0,184/' \
+    -e '/HeLa-S3/s/$/| color 0,119,158/' \
+    -e '/HepG2/s/$/| color 189,0,157/' \
+    -e '/HUVEC/s/$/| color 224,75,0/' \
+        encode3.lines > encode3.color.lines
+
+# buf size exceeded on linesToRa, so prune down to just composite subtracks
+
+linesToRa encode3.color.lines encode3.color.ra
+
+# concatenate to orig ra file
+
 #  reload cluster input table
 hgLoadSqlTab hg19 encode3RegTfbsClusterInput \
         ~/kent/src/hg/lib/clusterInputTrackTable5.sql clusters.inputs.tab
 
 # rename field in cluster input table
 # NOTE syntax change in MariaDb (now requires type)
 hgsql -e hg19 "alter table encRegTfbsClusteredInputs change treatment experiment varchar(255)"
 
 
 ###############
 # hg38
 
 cd ../hg38
 mkdir scoredPeaks
 cd peaks
@@ -409,30 +433,46 @@
 
 # Fix up subGroup members w/ punctuation and initial numbers:
 #       Peyer's_patch -> Peyers_patch
 #       NT2/D1 -> NT2_D1
 #       22Rv1 -> X22Rv1
 #       MM.1S -> MM_1S
 
 # rename tables (encode3Tfbs -> encTfChipPk)
 
 hgsql hg38 -e 'show tables like "encode3TfbsPk%"' > tables.old.txt
 sed -e 's/^/alter table /' -e 's/$/ rename to /' tables.old.txt > rename.1.sql
 sed -e 's/encode3TfbsPk/encTfChipPk/' tables.old.txt | paste rename.1.sql -  | \       
     sed 's/$/;/' > rename.sql
 hgsql hg38 < rename.sql
 
+# add colors, using DNase similarity track 
+raToLines encode3.ra encode3.lines
+
+# edit to leave only subtracks
+# generate edit script from DNase similarity .ra file
+
+cd ~/kent/src/hg/makeDb/trackDb/human/hg38
+csh addColors.csh encode3.lines > encode3.color.lines
+linesToRa encode3.color.lines encode3.color.ra
+
+# missed a few (punctuation diffs, etc.)
+csh addColors2.csh encode3.color.lines encode3.color2.lines
+linesToRa encode3.color2.lines encode3.color.ra
+
+# merge in to encode3.ra
+
 #  reload cluster input table
 hgLoadSqlTab hg38 encode3RegTfbsClusterInput \
         ~/kent/src/hg/lib/clusterInputTrackTable5.sql clusters.inputs.tab
 
 # rename field in cluster input table
 # NOTE syntax change in MariaDb (now requires type)
 hgsql -e hg19 "alter table encRegTfbsClusteredInputs change treatment experiment varchar(255)"
 
 ####################
 # Motifs (hg38) from Henry Pratt at Zlab
 # 2019-03-13
 
 # Notes from Henry:
 
 #The structure is mostly the same as the existing tables: canonical.tsv contains lists