# for emacs: -*- mode: sh; -*-

# This file describes browser build for the manPen1
# Chinese Pangolin - Manis pentadactyla

# DATE:   11-Aug-2014
# ORGANISM:       Manis pentadactyla
# TAXID:  143292
# ASSEMBLY LONG NAME:     M_pentadactyla-1.1.1
# ASSEMBLY SHORT NAME:    M_pentadactyla-1.1.1
# ASSEMBLY SUBMITTER:     Washington University (WashU)
# ASSEMBLY TYPE:  Haploid
# NUMBER OF ASSEMBLY-UNITS:       1
# ASSEMBLY ACCESSION:     GCA_000738955.1
# FTP-RELEASE DATE: 13-Aug-2014

# this build was run up with the older style genbank hierarchy

# picked up chrM from NC_016008.1

# Assembly Name:  M_pentadactyla-1.1.1
# Organism name:  Manis pentadactyla
# Taxid:          143292
# Submitter:      Washington University (WashU)
# Date:           2014-8-11
# BioSample:      SAMN01943338
# Assembly type:  haploid
# Release type:   major
# Assembly level: Scaffold
# Genome representation: full
# GenBank Assembly Accession: GCA_000738955.1 (species-representative latest)
#
## Assembly-Units:
## GenBank Unit Accession       RefSeq Unit Accession   Assembly-Unit name
## GCA_000738965.1              Primary Assembly


#############################################################################
# fetch sequence from new style download directory (DONE - 2014-08-13 - Hiram)
    mkdir -p /hive/data/genomes/manPen1/genbank
    cd /hive/data/genomes/manPen1/genbank

    rsync -L -a -P \
rsync://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Manis_pentadactyla/M_pentadactyla-1.1.1/ ./

    # measure what we have here:
    faSize Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fa.gz
    # 2204732179 bases (205794440 N's 1998937739 real 1998937739 upper
    #    0 lower) in 92772 sequences in 1 files
    # Total size: mean 23765.1 sd 56958.1
    #  min 200 (gi|671805652|gb|JPTV01206145.1|)
    #  max 1317973 (gi|672349252|gb|KN005617.1|) median 598

    # the newer hierarchy would have been from here:
    mkdir /hive/data/genomes/manPen1/newGenbank
    cd /hive/data/genomes/manPen1/newGenbank
    time rsync -L -a -P \
rsync://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Manis_pentadactyla/all_assembly_versions/GCA_000738955.1_M_pentadactyla-1.1.1/ ./

    # sent 318 bytes  received 1398411462 bytes  14054389.75 bytes/sec
    # total size is 1398238878  speedup is 1.00
    # real    1m39.586s

    faSize *.fna.gz
    # 2204732179 bases (205794440 N's 1998937739 real 1448708505 upper
    #    550229234 lower) in 92772 sequences in 1 files
    # Total size: mean 23765.1 sd 56958.1 min 200 (JPTV01206145.1)
    #    max 1317973 (KN005617.1) median 598
    #  %24.96 masked total, %27.53 masked real

export mitoAcc=NC_016008.1

wget -O ${mitoAcc}.fa \
 "http://www.ncbi.nlm.nih.gov/sviewer/viewer.fcgi?db=nuccore&dopt=fasta&sendto=on&id=$mitoAcc"

echo ">chrM" > chrM.fa
grep -v "^>" ${mitoAcc}.fa >> chrM.fa

export mSize=`faCount chrM.fa | grep total | awk '{print $2}'`

/bin/echo -e "chrM\t1\t$mSize\t1\tF\t$mitoAcc\t1\t$mSize\t+" > chrM.agp

#############################################################################
# fixup to UCSC naming scheme (DONE - 2014-08-13 - Hiram)
    mkdir /hive/data/genomes/manPen1/ucsc
    cd /hive/data/genomes/manPen1/ucsc

    # three duplicate contigs were removed:
# JPTV01168508
# JPTV01172399
# JPTV01172711

    # older style name mangling, simply remove the .1 on all the names
    # and the three duplicates

zcat ../genbank/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \
    | egrep -v "^#|JPTV01168508|JPTV01172399|JPTV01172711" \
       | sed -e 's/\.1//;' > chrUn.agp

zcat ../genbank/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fa.gz \
    | sed -e 's/.1. Manis .*//; s/gi.[0-9][0-9]*.gb.//;' \
       | faSomeRecords -exclude stdin dup.list stdout | gzip -c > chrUn.fa.gz

    zcat chrUn.fa.gz chrM.fa.gz | grip -c > manPen1.ucsc.fa.gz
    zcat chrM.agp chrM.agp > manPen1.ucsc.agp

    # verify these two files are compatible:
    zcat chrUn.fa.gz chrM.fa.gz | faToTwoBit stdin.manPen1.fa test.2bit
    checkAgpAndFa manPen1.ucsc.agp test.2bit 2>&1 | tail
    # All AGP and FASTA entries agree - both files are valid
    rm -f test.2bit

#############################################################################
# photo (DONE - 2015-06-10 - Hiram)
    mkdir /hive/data/genomes/manPen1/photo
    cd /hive/data/genomes/manPen1/photo
    # found photo at flickr without restrictions
    # from user dotpolka: https://www.flickr.com/photos/dotpolka/
    # aka: https://twitter.com/dotpolka Caroline Härdter
    wget --timestamping \
https://farm1.staticflickr.com/39/78362867_c7b9c42166_o_d.jpg

    convert -quality 100 -crop 1268x2684+192+256 \
        78362867_c7b9c42166_o_d.jpg pangolinCrop.jpg
    convert -quality 80 -geometry 300x400 \
        pangolinCrop.jpg Manis_pentadactyla.jpg

    # check that Manis_pentadactyla.jpg file into source tree
    # src/hg/htdocs/images/ and copy to /usr/local/apache/htdocs/images/

#############################################################################
#  Initial database build (DONE - 2014-08-14 - Hiram)

    cd /hive/data/genomes/manPen1
    cat << '_EOF_' > manPen1.config.ra
# Config parameters for makeGenomeDb.pl:
db manPen1
clade mammal
genomeCladePriority 31
scientificName Manis pentadactyla
commonName Chinese pangolin
assemblyDate Aug. 2014
assemblyLabel Washington University
assemblyShortLabel M_pentadactyla-1.1.1
orderKey 3355
# mitoAcc included in source fasta prepared above
# mitoAcc NC_016008.1
mitoAcc none
fastaFiles /hive/data/genomes/manPen1/ucsc/manPen1.ucsc.fa.gz
agpFiles /hive/data/genomes/manPen1/ucsc/manPen1.ucsc.agp
# qualFiles none
dbDbSpeciesDir pangolin
photoCreditURL https://www.flickr.com/people/dotpolka/
photoCreditName Caroline Härdter/flicker
ncbiGenomeId 10694
ncbiAssemblyId 203041
ncbiAssemblyName M_pentadactyla-1.1.1
ncbiBioProject 20331
genBankAccessionID GCA_000738955.1
taxId 143292
'_EOF_'
    # << happy emacs

    # these steps were performed manually before the manPen1.config.ra file
    # was constructed.

    # verify sequence and AGP are OK:
    time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \
         -stop=agp manPen1.config.ra) > agp.log 2>&1
    # *** All done!  (through the 'agp' step)
    # real    66m20.834s

    # then finish it off:
    time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \
       -fileServer=hgwdev -continue=db manPen1.config.ra) > db.log 2>&1
    # real    35m25.469s

    # did run this up with the new manPen1.config.ra file: 2015-06-10

    makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileserver=hgwdev \
       -continue=trackDb manPen1.config.ra > trackDb.debug.log 2>&1

    # check in the trackDb files created and add to trackDb/makefile
    # and clean up:
    rm -fr TemporaryTrackDbCheckout/

##########################################################################
# running repeat masker (DONE - 2014-08-15 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/repeatMasker
    cd /hive/data/genomes/manPen1/bed/repeatMasker
    time (doRepeatMasker.pl -buildDir=`pwd` \
      -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev -useRMBlastn \
       -smallClusterHub=ku manPen1) > do.log 2>&1
    # real    558m51.310s

    cat faSize.rmsk.txt
    # 2204741241 bases (205794437 N's 1998946804 real 1343043070 upper
    #    655903734 lower) in 92770 sequences in 1 files
    # Total size: mean 23765.7 sd 56958.6 min 200 (JPTV01206145)
    #   max 1317973 (KN005617) median 598
    # %29.75 masked total, %32.81 masked real

    egrep -i "versi|relea" do.log
    # RepeatMasker version open-4.0.5
    #    January 31 2015 (open-4-0-5) version of RepeatMasker
    # CC   RELEASE 20140131;

    time featureBits -countGaps manPen1 rmsk
    # 655984124 bases of 2204741241 (29.753%) in intersection
    # real    1m0.941s

    # why is it different than the faSize above ?
    # because rmsk masks out some N's as well as bases, the count above
    #   separates out the N's from the bases, it doesn't show lower case N's

##########################################################################
# running simple repeat (DONE - 2014-08-14 - Hiram)

    mkdir /hive/data/genomes/manPen1/bed/simpleRepeat
    cd /hive/data/genomes/manPen1/bed/simpleRepeat
    time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \
        -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \
        manPen1) > do.log 2>&1
    # real    39m22.135s

    cat fb.simpleRepeat
    # 43106952 bases of 1999066070 (2.156%) in intersection

    # using the Window Masker result:
    cd /hive/data/genomes/manPen1
    twoBitMask bed/repeatMasker/manPen1.rmsk.2bit \
      -add bed/simpleRepeat/trfMask.bed  manPen1.2bit
    #   you can safely ignore the warning about fields >= 13
    twoBitToFa manPen1.2bit stdout | faSize stdin > faSize.manPen1.2bit.txt
    cat faSize.manPen1.2bit.txt
    # 2204741241 bases (205794437 N's 1998946804 real 1342683024 upper
    #    656263780 lower) in 92770 sequences in 1 files
    # Total size: mean 23765.7 sd 56958.6 min 200 (JPTV01206145)
    #    max 1317973 (KN005617) median 598
    # %29.77 masked total, %32.83 masked real

    rm /gbdb/manPen1/manPen1.2bit
    ln -s `pwd`/manPen1.2bit /gbdb/manPen1/manPen1.2bit

##########################################################################
# CREATE MICROSAT TRACK (DONE - 2015-06-22 - Hiram)
     ssh hgwdev
     mkdir /cluster/data/manPen1/bed/microsat
     cd /cluster/data/manPen1/bed/microsat
     awk '($5==2 || $5==3) && $6 >= 15 && $8 == 100 && $9 == 0 {printf("%s\t%s\t%s\t%dx%s\n", $1, $2, $3, $6, $16);}' \
	../simpleRepeat/simpleRepeat.bed > microsat.bed
    hgLoadBed manPen1 microsat microsat.bed
    #	Loaded 13365 elements of size 4

##########################################################################
## WINDOWMASKER (DONE - 2014-08-16 - Hiram)

    mkdir /hive/data/genomes/manPen1/bed/windowMasker
    cd /hive/data/genomes/manPen1/bed/windowMasker
    time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \
        -dbHost=hgwdev manPen1) > do.log 2>&1
    # real    252m52.324s

    # Masking statistics
    cat faSize.manPen1.cleanWMSdust.txt
    # 2204741241 bases (205794437 N's 1998946804 real 1438998713 upper
    #    559948091 lower) in 92770 sequences in 1 files
    # Total size: mean 23765.7 sd 56958.6 min 200 (JPTV01206145)
    #    max 1317973 (KN005617) median 598
    # %25.40 masked total, %28.01 masked real

    cat fb.manPen1.rmsk.windowmaskerSdust.txt
    # 276979736 bases of 2204741241 (12.563%) in intersection

##########################################################################
# cpgIslands - (DONE - 2014-08-19 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/cpgIslands
    cd /hive/data/genomes/manPen1/bed/cpgIslands
    time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \
      -workhorse=hgwdev -smallClusterHub=ku manPen1) > do.log 2>&1 &
    # real    169m34.228s

    cat fb.manPen1.cpgIslandExt.txt
    # 11640846 bases of 1999066070 (0.582%) in intersection

##############################################################################
# cpgIslands on UNMASKED sequence (DONE - 2014-08-20 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/cpgIslandsUnmasked
    cd /hive/data/genomes/manPen1/bed/cpgIslandsUnmasked

    time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \
       -tableName=cpgIslandExtUnmasked \
          -maskedSeq=/hive/data/genomes/manPen1/manPen1.unmasked.2bit \
             -workhorse=hgwdev -smallClusterHub=ku manPen1) > do.log 2>&1
    # real    360m14.919s

    cat fb.manPen1.cpgIslandExtUnmasked.txt
    # 12785608 bases of 1999066070 (0.640%) in intersection

#############################################################################
# cytoBandIdeo - (DONE - 2015-06-10 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/cytoBand
    cd /hive/data/genomes/manPen1/bed/cytoBand
    makeCytoBandIdeo.csh manPen1

#########################################################################
# augustus - (DONE - 2015-05-28 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/augustus
    cd /hive/data/genomes/manPen1/bed/augustus

export db="manPen1"
export species="human"
doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \
  -species=$species -dbHost=hgwdev -utr -workhorse=hgwdev $db

    cat fb.manPen1.augustusGene.txt
    #  49832677 bases of 1999066070 (2.493%) in intersection

#########################################################################
# genscan - (DONE - 2014-08-22 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/genscan
    cd /hive/data/genomes/manPen1/bed/genscan
    time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \
      -bigClusterHub=ku manPen1) > do.log 2>&1 &
    # real    122m8.890s

    cat fb.manPen1.genscan.txt
    #   55603181 bases of 1999066070 (2.781%) in intersection

    cat fb.manPen1.genscanSubopt.txt
    #   62862446 bases of 1999066070 (3.145%) in intersection

#########################################################################
# ucscToINSDC table/track (DONE - 2015-06-22 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/ucscToINSDC
    cd /hive/data/genomes/manPen1/bed/ucscToINSDC

    # check for chrM accession:
    grep chrM ../../manPen1.agp
# chrM    1       16578   1       F       NC_016008       1       16578   +

    # use that accession as an argument to this command

    ~/kent/src/hg/utils/automation/ucscToINSDC.sh \
         ../../genbank/Primary_Assembly NC_016008

    awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \
         | sort > name.coordinate.tab
    # do not need the v1 in this case:
    sed -e 's/v1//' ucscToINSDC.txt \
      | join name.coordinate.tab - | tr '[ ]' '[\t]' \
         > ucscToINSDC.bed
    # verify all names are coming through, should be same line count:
    wc -l *
    #    92770 name.coordinate.tab
    #    92770 ucscToINSDC.bed
    #    92773 ucscToINSDC.txt

    # three missing ones are the dups that UCSC removed:
     cat ../../ucsc/dup.list
JPTV01168508
JPTV01172399
JPTV01172711

    cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1
    # 12
    # use the 12 in this sed
    sed -e "s/21/12/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \
         | hgLoadSqlTab manPen1 ucscToINSDC stdin ucscToINSDC.bed
    checkTableCoords manPen1
    # should cover %100 entirely:
    featureBits -countGaps manPen1 ucscToINSDC
    # 2204741241 bases of 2204741241 (100.000%) in intersection

#########################################################################
# Create kluster run files (DONE - 2015-06-10 - Hiram)

    cd /hive/data/genomes/manPen1
    # numerator is manPen1 gapless bases "real" as reported by:
    head -1 faSize.manPen1.2bit.txt
# 2204741241 bases (205794437 N's 1998946804 real 1342683024 upper
#    656263780 lower) in 92770 sequences in 1 files

    # numerator is 'real' base count
    # denominator is hg19 gapless bases as reported by:
    #   featureBits -noRandom -noHap hg19 gap
    #     234344806 bases of 2861349177 (8.190%) in intersection
    # 1024 is threshold used for human -repMatch:
    calc \( 1998946804 / 2861349177 \) \* 1024
    # ( 1998946804 / 2861349177 ) * 1024 = 715.369359

    # ==> use -repMatch=700 according to size scaled down from 1024 for human.
    #   and rounded down to nearest 50
    cd /hive/data/genomes/manPen1
    time blat manPen1.2bit \
         /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/manPen1.11.ooc \
        -repMatch=700
    # Wrote 25574 overused 11-mers to jkStuff/manPen1.11.ooc
    #   real    0m33.809s

    # there are no non-bridged gaps, do not need to do this
    #   check non-bridged gaps to see what the typical size is:
#     hgsql -N -e 'select * from gap where bridge="no" order by size;' manPen1
#       | ave -tableOut -col=7 stdin
# # min Q1 median Q3 max mean N sum stddev
# 50076 58368.8 70128 100495 1.07816e+07 178173 670 1.19376e+08 672006

    # note the minimum non-bridged gap size is 50,076

#  gapToLift -verbose=2 -minGap=50000 manPen1 jkStuff/manPen1.nonBridged.lft \
#         -bedFile=jkStuff/manPen1.nonBridged.bed

#    hgsql -N \
#        -e 'select * from gap where bridge="no" order by size;' manPen1 \
#        | ave -col=7 stdin

    # not needed:
#    gapToLift -verbose=2 -minGap=100 bosTau7 jkStuff/nonBridged.lft \
#        -bedFile=jkStuff/nonBridged.bed

    # survey sizes:
    n50.pl chrom.sizes
#       reading: chrom.sizes
#       contig count: 92770, total size: 2204741241, one half size: 1102370620
# cumulative    N50 count       contig  contig size
# 1102318455      5267    KN010628        117930
# 1102370620 one half size
# 1102436375      5268    KN010629        117920

#############################################################################
# GENBANK AUTO UPDATE (DONE - 2015-06-10 - Hiram)
    ssh hgwdev
    cd $HOME/kent/src/hg/makeDb/genbank
    git pull
    # /cluster/data/genbank/data/organism.lst shows one mRNA:
    # #organism       mrnaCnt estCnt  refSeqCnt
    # Manis javanica  1       0       0

    # edit etc/genbank.conf to add manPen1 just before bosTau2

# manPen1 (Pangolin) 92770 contigs
manPen1.serverGenome = /hive/data/genomes/manPen1/manPen1.2bit
manPen1.clusterGenome = /hive/data/genomes/manPen1/manPen1.2bit
manPen1.ooc = /hive/data/genomes/manPen1/jkStuff/manPen1.11.ooc
manPen1.lift = no
manPen1.refseq.mrna.xeno.pslCDnaFilter    = ${lowCover.refseq.mrna.xeno.pslCDnaFilter}
manPen1.genbank.mrna.native.pslCDnaFilter = ${lowCover.genbank.mrna.native.pslCDnaFilter}
manPen1.genbank.mrna.xeno.pslCDnaFilter   = ${lowCover.genbank.mrna.xeno.pslCDnaFilter}
manPen1.genbank.est.native.pslCDnaFilter  = ${lowCover.genbank.est.native.pslCDnaFilter}
manPen1.genbank.est.xeno.pslCDnaFilter    = ${lowCover.genbank.est.xeno.pslCDnaFilter}
manPen1.downloadDir = manPen1
manPen1.refseq.mrna.native.load  = no
manPen1.refseq.mrna.xeno.load = yes
manPen1.genbank.mrna.xeno.load = no
manPen1.genbank.mrna.native.load = no
manPen1.genbank.est.native.load = no
manPen1.perChromTables = no

    git commit -m "Added manPen1/Pangolin; refs #15515" etc/genbank.conf
    git push
    # update /cluster/data/genbank/:
    make etc-update

    # Edit src/lib/gbGenome.c to add new species.
    git commit -m "Added manPen Pangolin/Manis pentadactyla (javanica) refs #15515" \
       src/lib/gbGenome.c
    git push
    make install-server

    screen      #  control this business with a screen since it takes a while
    cd /cluster/data/genbank

    time ./bin/gbAlignStep -initial manPen1
    # logFile: var/build/logs/2015.06.10-14:04:03.manPen1.initalign.log
    # real    34m30.608s

    #   To re-do, rm the dir first:
    #     /cluster/data/genbank/work/initial.manPen1

    # load database when finished
    ssh hgwdev
    cd /cluster/data/genbank
    time ./bin/gbDbLoadStep -drop -initialLoad manPen1
    # logFile: var/dbload/hgwdev/logs/2015.06.15-09:40:01.manPen1.dbload.log
    # real    6m45.556s

    # enable daily alignment and update of hgwdev
    cd ~/kent/src/hg/makeDb/genbank
    git pull
    # add manPen1 to:
    #   vi etc/align.dbs etc/hgwdev.dbs
    git commit -m "Added manPen1 - Pangolin/Manis pentadactyla (javanica)refs #15515" \
      etc/align.dbs etc/hgwdev.dbs
    git push
    make etc-update

#########################################################################
# ucscToINSDC table/track (DONE - 2015-03-10 - Hiram)
    mkdir /hive/data/genomes/manPen1/bed/ucscToINSDC
    cd /hive/data/genomes/manPen1/bed/ucscToINSDC

    ~/kent/src/hg/utils/automation/ucscToINSDC.sh \
       ../../genbank/GCA_*assembly_structure/Primary_Assembly NC_012346.1

    awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \
         | sort > name.coordinate.tab
    join name.coordinate.tab ucscToINSDC.txt | tr '[ ]' '[\t]' \
         > ucscToINSDC.bed
    # verify all names are coming through, should be same line count:
    wc -l *
    #   450182 name.coordinate.tab
    #   450182 ucscToINSDC.bed
    #   450182 ucscToINSDC.txt

    cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1
    # 14
    # use the 14 in this sed
    sed -e "s/21/14/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \
         | hgLoadSqlTab manPen1 ucscToINSDC stdin ucscToINSDC.bed
    checkTableCoords manPen1
    # should cover %100 entirely:
    featureBits -countGaps manPen1 ucscToINSDC
    # 1192725744 bases of 1192725744 (100.000%) in intersection

#########################################################################
# fixup search rule for assembly track/gold table (DONT - 2017-03-28 - Hiram)
    cd ~/kent/src/hg/makeDb/trackDb/pangolin/manPen1

    # preview prefixes and suffixes:
    hgsql -N -e "select frag from gold;" manPen1 \
      | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c
#  230927 JPTV.1
#       1 NC_

    # implies a rule: '[JN][CP][T_][V0-9]+(\.[0-9]+)?'

    # verify this rule will find them all and eliminate them all:
    hgsql -N -e "select frag from gold;" manPen1 | wc -l
    # 230928

    hgsql -N -e "select frag from gold;" manPen1 \
       | egrep -e '[JN][CP][T_][V0-9]+(\.[0-9]+)?' | wc -l
    # 230928

    hgsql -N -e "select frag from gold;" manPen1 \
       | egrep -v -e '[JN][CP][T_][V0-9]+(\.[0-9]+)?' | wc -l
    # 0

    # hence, add to trackDb/chicken/manPen1/trackDb.ra
searchTable gold
shortCircuit 1
termRegex [JN][CP][T_][V0-9]+(\.[0-9]+)?
query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%'
searchPriority 8

    # verify searches work in the position box

############################################################################
#  BLATSERVERS ENTRY (DONE - 2017-03-22 - Hiram)
#	After getting a blat server assigned by the Blat Server Gods,
    ssh hgwdev

    hgsql -e 'INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
	VALUES ("manPen1", "blat1b", "17878", "1", "0"); \
	INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
	VALUES ("manPen1", "blat1b", "17879", "0", "1");' \
	    hgcentraltest
    #	test it with some sequence

############################################################################
## reset default position to MAOA gene (Monoamine Oxidase A from monDom5)
##  (DONE - 2017-03-22 - Hiram)

    ssh hgwdev
    hgsql -e 'update dbDb set defaultPos="KN007563:109071-158662"
	where name="manPen1";' hgcentraltest

############################################################################
# all.joiner update, downloads and in pushQ - (DONE 2014-10-21 - Steve)
    cd $HOME/kent/src/hg/makeDb/schema
    # fixup all.joiner until this is a clean output
    git commit -m "Added manPen1 - Chinese Pangolin/Manis pentadactyl arefs #15515" \
       all.joiner
    joinerCheck -database=manPen1 -tableCoverage all.joiner
    joinerCheck -database=manPen1 -times all.joiner
    joinerCheck -database=manPen1 -keys all.joiner

    cd /hive/data/genomes/manPen1
    time makeDownloads.pl manPen1 > downloads.log 2>&1
    # real    14m28.068s

    #   now ready for pushQ entry
    mkdir /hive/data/genomes/manPen1/pushQ
    cd /hive/data/genomes/manPen1/pushQ
    time makePushQSql.pl manPen1 > manPen1.pushQ.sql 2> stderr.out
    # real    5m58.381s

    #   check for errors in stderr.out, some are OK, e.g.:
    # WARNING: hgwdev does not have /gbdb/manPen1/wib/gc5Base.wib
    # WARNING: hgwdev does not have /gbdb/manPen1/wib/quality.wib
    # WARNING: hgwdev does not have /gbdb/manPen1/bbi/qualityBw/quality.bw
    # WARNING: manPen1 does not have seq
    # WARNING: manPen1 does not have extFile
    # WARNING: manPen1 does not have estOrientInfo
    # WARNING: manPen1 does not have mrnaOrientInfo

    #   copy it to hgwbeta
    scp -p manPen1.pushQ.sql qateam@hgwbeta:/tmp
    ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/manPen1.pushQ.sql"
    #   in that pushQ entry walk through each entry and see if the
    #   sizes will set properly

#########################################################################