src/hg/makeDb/doc/hg19.txt 1.103
1.103 2010/04/21 20:57:24 chinhli
Add sno/,iRNA track
Index: src/hg/makeDb/doc/hg19.txt
===================================================================
RCS file: /projects/compbio/cvsroot/kent/src/hg/makeDb/doc/hg19.txt,v
retrieving revision 1.102
retrieving revision 1.103
diff -b -B -U 4 -r1.102 -r1.103
--- src/hg/makeDb/doc/hg19.txt 20 Apr 2010 17:34:31 -0000 1.102
+++ src/hg/makeDb/doc/hg19.txt 21 Apr 2010 20:57:24 -0000 1.103
@@ -9694,5 +9694,94 @@
nice gzip snp131Simple.bed snp131ExcludeIds.txt snp131ForLiftOver.bed
rm -r run*/split tmp.txt *.orthoGlom.txt
+
+##############################################################################
+# RE-BUILD sno/miRNA TRACK (DONE - 04-20-2010 - Chin)
+
+ # The data in this track is out of date so update the track.
+ mkdir -p /hive/data/genomes/hg19/bed/wgRna-2010-04-20
+ cd /hive/data/genomes/hg19/bed/wgRna-2010-04-20
+
+ # Download GFF file of latest miRNA annotations from miRBase at the
+ # ftp://mirbase.org/pub/mirbase/CURRENT/. This is Release 14.0
+ # (September, 2009)
+ wget --timestamping \
+ ftp://mirbase.org/pub/mirbase/CURRENT/genomes/hsa.gff
+ # Re-format, need to add "chr" to the beginning of each line.
+ sed -e 's/^/chr/' hsa.gff > hsMirBaseFormat.gff
+ # Remove extra "chr" in comment lines
+ perl -pi.bak -e 's/chr#/#/' hsMirBaseFormat.gff
+ # Change chrMT to chrM
+ perl -pi.bak -e 's/chrMT/chrM/' hsMirBaseFormat.gff
+ # Remove all but ID name in last field
+ sed -e 's/\";//g' hsMirBaseFormat.gff | sed -e 's/ID=\"//g' \
+ | sed -e 's/ACC=\"MI[0-9]*\s//' > hsMirBaseFormatIdOnly.gff
+
+ # use score 906 for + strand and 480 for - strand. This will show
+ # up black on the track for + strand and grey for - strand.
+ # Starts appear to be 1-based when compared to miRNAs in current
+ # track
+ # and those in Ensembl.
+ # Confirmed with Sam Griffith-Jones (one of the authors of miRBase,
+ # sam.griffith-jones@manchester.ac.uk) that these GFF coordinates
+ # are 1-based.
+ # Also add thickStart and thickEnd columns and "miRNA" for type.
+ awk 'BEGIN {FS="\t"} {OFS="\t"} \
+ {if ($0 !~ /#/ && $7 == "+") \
+ print $1, $4-1, $5, $9, 960, $7, 0, 0, "miRNA"; \
+ else if ($0 !~ /#/ && $7 == "-") \
+ print $1, $4-1, $5, $9, 480, $7, 0, 0, "miRNA";}' \
+ hsMirBaseFormatIdOnly.gff > hsMirBaseFormatIdOnly.bed
+
+ # 2010-04-21
+ # Down load the current snoRNABase coordinates (version 3, based on hg19)
+ # from
+ # http://www-snorna.biotoul.fr/coordinates.php
+ # to
+ # /hive/data/genomes/hg19/bed/wgRna-2010-04-20/snoRNABaseVer3Coords.xls
+
+ cd /hive/data/genomes/hg19/bed/wgRna-2010-04-20/
+ cp snoRNABaseVer3Coords.xls snoRNABaseVer3Coords.txt
+ # remove the header line (column title).
+ # remove all the quotes surrounding characters field
+ perl -pi.bak -e 's/\"//g' snoRNABaseVer3Coords.txt
+ # Reformat to BED format with thickStart and thickEnd set to 0.
+ awk 'BEGIN {FS="\t"} {OFS="\t"} \
+ {if ($4 == "+") \
+ print $1, $2-1, $3, $5, 960, $4, 0, 0,$6; \
+ else if ($4 == "-") \
+ print $1, $2-1, $3, $5, 480, $4, 0, 0,$6;}' \
+ snoRNABaseVer3Coords.txt > snoRNABaseVer3Coords.bed
+ # Merge the miRNA and snoRNA files together
+ cat hsMirBaseFormatIdOnly.bed snoRNABaseVer3Coords.bed \
+ > wgRna20100420.bed
+ # Create and load wgRna
+ cp -p /cluster/home/chinhli/kent/src/hg/lib/wgRna.sql wgRna.sql
+ hgLoadBed -sqlTable=wgRna.sql hg19 wgRna wgRna20100420.bed
+ # Reading wgRna20100420.bed
+ # Loaded 1152 elements of size 9
+ # Sorted
+ # Creating table definition for wgRna
+ # Saving bed.tab
+ # Loading hg19
+
+ # Clean up
+ rm *.bak
+
+ # some details about this track:
+ hgsql -e "select count(*) from wgRna;" hg19
+ # 1152
+ # contain 4 types:
+ cat wgRna20100420.bed | awk '{print $9}' | sort | uniq
+ hgsql -e "select type, count(*) from wgRna group by type;" hg19
+ # CDBox 269
+ # HAcaBox 112
+ # miRNA 750
+ # scaRna 21
+ featureBits hg19 wgRna
+ # 107878 bases of 2897316137 (0.004%) in intersection
+
+
+
############################################################################