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

# This file describes browser build for the panTro6

#########################################################################
# reuse photograph from panTro5 previous versions
#    (DONE - 2018-03-23 - Hiram)

mkdir /hive/data/genomes/panTro6
cd /hive/data/genomes/panTro6
cp -p ../panTro5/photoReference.txt .

cat photoReference.txt

photoCreditURL  http://www.smm.org/
photoCreditName Science Museum of Minnesota

#########################################################################
#  Initial steps (DONE - 2018-03-23 - Hiram)

# To start this initialBuild.txt document, from a previous assembly document:

mkdir ~/kent/src/hg/makeDb/doc/panTro6
cd ~/kent/src/hg/makeDb/doc/panTro6

# best to use a most recent document since it has the latest features and
# procedures:
sed -e 's/felCat9/panTro6/g; s/FelCat9/PanTro6/g; s/DONE/TBD/g;' ../felCat9/initialBuild.txt > initialBuild.txt

mkdir /hive/data/genomes/panTro6/refseq
cd /hive/data/genomes/panTro6/refseq

time rsync -L -a -P \
rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Pan_troglodytes/all_assembly_versions/GCF_002880755.1_Clint_PTRv2/ ./

# sent 3106 bytes  received 4399985960 bytes  18219416.42 bytes/sec
# total size is 4399437472  speedup is 1.00

# real    4m0.842s

# check assembly size for later reference:

faSize G*_PTRv2_genomic.fna.gz

# 3050398082 bases (31805092 N's 3018592990 real 1820051887 upper
#	1198541103 lower) in 4346 sequences in 1 files
# Total size: mean 701886.4 sd 9434013.2 min 223 (NW_019937062.1)
#	max 224244399 (NC_036879.1) median 29187
# %39.29 masked total, %39.71 masked real

# this information is from the top of
#    panTro6/refseq/GCF_002880755.1_Clint_PTRv2_assembly_report.txt


# Assembly name:  Clint_PTRv2
# Organism name:  Pan troglodytes (chimpanzee)
# Isolate:  Yerkes chimp pedigree #C0471 (Clint)
# Sex:  male
# Taxid:          9598
# BioSample:      SAMN06272697
# BioProject:     PRJNA369439
# Submitter:      University of Washington
# Date:           2018-1-19
# Assembly type:  haploid
# Release type:   major
# Assembly level: Chromosome
# Genome representation: full
# WGS project:    NBAG03
# Expected final version: yes
# Genome coverage: 124x
# RefSeq assembly accession: GCF_002880755.1
#
## Assembly-Units:
## GenBank Unit Accession	RefSeq Unit Accession	Assembly-Unit name
## 	GCF_002880765.1	Primary Assembly
## 	GCF_000001485.1	non-nuclear

#############################################################################
# establish config.ra file (DONE - Hiram - 2018-03-23)
    # arguments here are: <db> <clade> <trackDbDir> <assembly_report.txt>
    cd /hive/data/genomes/panTro6
    $HOME/kent/src/hg/utils/automation/prepConfig.pl panTro6 mammal \
        chimp ./refseq/*_assembly_report.txt > panTro6.config.ra

    # compare to ../panTro5 to see what might need to be fixed up:
    diff panTro6.config.ra ../panTro5/panTro5.config.ra  | less

    # reset commonName Chimpanzee to: commonName Chimp
    # reset orderKey 3337 to: orderKey 3325
    # reset ncbiGenomeId 325 to: ncbiGenomeId 202
    # verify it looks sane
    cat panTro6.config.ra
# config parameters for makeGenomeDb.pl:
db panTro6
clade mammal
genomeCladePriority 35
scientificName Pan troglodytes
commonName Chimp
assemblyDate Jan. 2018
assemblyLabel University of Washington
assemblyShortLabel Clint_PTRv2
orderKey 3325
# mitochondrial sequence included in refseq release
# mitoAcc NC_001643.1
mitoAcc none
fastaFiles /hive/data/genomes/panTro6/ucsc/*.fa.gz
agpFiles /hive/data/genomes/panTro6/ucsc/*.agp
# qualFiles none
dbDbSpeciesDir chimp
photoCreditURL  http://www.smm.org/
photoCreditName Science Museum of Minnesota
ncbiGenomeId 202
ncbiAssemblyId 1642151
ncbiAssemblyName Clint_PTRv2
ncbiBioProject 369439
ncbiBioSample SAMN06272697
genBankAccessionID GCF_002880755.1
taxId 9598

#############################################################################
# setup UCSC named files (DONE - 2018-03-23 - Hiram)

    mkdir /hive/data/genomes/panTro6/ucsc
    cd /hive/data/genomes/panTro6/ucsc

    # check for duplicate sequences:
    time faToTwoBit -noMask ../refseq/G*_PTRv2_genomic.fna.gz refseq.2bit
    #  real    1m17.463s

    twoBitDup refseq.2bit
    # no output is a good result, otherwise, would have to eliminate duplicates
    # the scripts creating the fasta here will be using this refseq.2bit file

    time ~/kent/src/hg/utils/automation/ucscCompositeAgp.pl \
       ../refseq/G*_PTRv2_genomic.fna.gz \
          ../refseq/G*_PTRv2_assembly_structure/Primary_Assembly
# NC_006492.4 chrY
# NC_036879.1 chr1
# NC_036880.1 chr2A
# NC_036881.1 chr2B
# NC_036882.1 chr3
# NC_036883.1 chr4
# NC_036884.1 chr5
# NC_036885.1 chr6
# NC_036886.1 chr7
# NC_036887.1 chr8
# NC_036888.1 chr9
# NC_036889.1 chr10
# NC_036890.1 chr11
# NC_036891.1 chr12
# NC_036892.1 chr13
# NC_036893.1 chr14
# NC_036894.1 chr15
# NC_036895.1 chr16
# NC_036896.1 chr17
# NC_036897.1 chr18
# NC_036898.1 chr19
# NC_036899.1 chr20
# NC_036900.1 chr21
# NC_036901.1 chr22
# NC_036902.1 chrX
# real    16m9.446s
    # unplaced sequences
    time ~/kent/src/hg/utils/automation/unplacedWithChroms.pl \
       ../refseq/*_assembly_structure/Primary_Assembly
# processed 4316 sequences into chrUn.fa.gz
# real    0m44.310s

    # unlocalized sequences
    time ~/kent/src/hg/utils/automation/unlocalizedWithChroms.pl \
       ../refseq/*_assembly_structure/Primary_Assembly
# 16
# 5
# processed 4 sequences into chr*_random.gz 2 files
# real    0m6.591s

    # bash syntax here
    mitoAcc=`grep "^# mitoAcc" ../panTro6.config.ra | awk '{print $NF}'`
    printf "# mitoAcc %s\n" "$mitoAcc"
# mitoAcc NC_001643.1

    zcat \
  ../refseq/*_assembly_structure/non-nuclear/assem*/AGP/chrMT.comp.agp.gz \
     | grep -v "^#" | sed -e "s/^$mitoAcc/chrM/;" > chrM.agp

    printf ">chrM\n" > chrM.fa
    twoBitToFa -noMask refseq.2bit:$mitoAcc stdout | grep -v "^>" >> chrM.fa
    gzip chrM.fa

    # verify fasta and AGPs agree
    time faToTwoBit *.fa.gz test.2bit
    # real    1m22.353s

    time cat *.agp | checkAgpAndFa stdin test.2bit 2>&1 | tail -4
    # All AGP and FASTA entries agree - both files are valid
    # real    0m10.242s

    # and no sequence lost from orginal:
    twoBitToFa test.2bit stdout | faSize stdin
# 3050398082 bases (31805092 N's 3018592990 real 3018592990 upper 0 lower)
#	in 4346 sequences in 1 files
# Total size: mean 701886.4 sd 9434013.2 min 223 (chrUn_NW_019937062v1)
#	max 224244399 (chr1) median 29187

    # same numbers as above
# 3050398082 bases (31805092 N's 3018592990 real 1820051887 upper
#	1198541103 lower) in 4346 sequences in 1 files
# Total size: mean 701886.4 sd 9434013.2 min 223 (NW_019937062.1)
#	max 224244399 (NC_036879.1) median 29187

    # no longer need these temporary 2bit files
    rm refseq.2bit test.2bit

#############################################################################
#  Initial database build (DONE - 2018-03-23 - Hiram)

    cd /hive/data/genomes/panTro6
    # verify sequence and AGP are OK:
    time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \
         -stop=agp panTro6.config.ra) > agp.log 2>&1
    # real    2m43.743s

    # verify there was no error in that step:
    tail agp.log
    #  *** All done!  (through the 'agp' step)

    # then finish it off:
    time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \
       -fileServer=hgwdev -continue=db panTro6.config.ra) > db.log 2>&1
    # real    24m48.575s

    # verify gaps are all there:
    twoBitInfo -nBed panTro6.unmasked.2bit stdout | awk '{print $3-$2}' \
	| ave stdin | sed -e 's/^/# /;'
# Q1 100.000000
# median 499.000000
# Q3 50212.500000
# average 44482.646154
# min 10.000000
# max 621330.000000
# count 715
# total 31805092.000000

    hgsql -e 'select chromEnd-chromStart from gap;' panTro6 | ave stdin | sed -e 's/^/# /;'
# Q1 100.000000
# median 499.000000
# Q3 50212.500000
# average 44482.646154
# min 10.000000
# max 621330.000000
# count 715
# total 31805092.000000


    # check in the trackDb files created in TemporaryTrackDbCheckout/
    #    and add panTro6 to trackDb/makefile

    # temporary symlink until masked sequence is available
    cd /hive/data/genomes/panTro6
    ln -s `pwd`/panTro6.unmasked.2bit /gbdb/panTro6/panTro6.2bit

##############################################################################
# cpgIslands on UNMASKED sequence (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/cpgIslandsUnmasked
    cd /hive/data/genomes/panTro6/bed/cpgIslandsUnmasked

    time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \
       -tableName=cpgIslandExtUnmasked \
          -maskedSeq=/hive/data/genomes/panTro6/panTro6.unmasked.2bit \
             -workhorse=hgwdev -smallClusterHub=ku panTro6) > do.log 2>&1
    # real    6m17.200s

    cat fb.panTro6.cpgIslandExtUnmasked.txt
    # 29545528 bases of 3018592990 (0.979%) in intersection

#############################################################################
# cytoBandIdeo - (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/cytoBand
    cd /hive/data/genomes/panTro6/bed/cytoBand
    makeCytoBandIdeo.csh panTro6

#############################################################################
# gapOverlap (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/gapOverlap
    cd /hive/data/genomes/panTro6/bed/gapOverlap
    time (doGapOverlap.pl \
	-twoBit=/hive/data/genomes/panTro6/panTro6.unmasked.2bit panTro6 ) \
	> do.log 2>&1
    # real    14m5.164s

    cat fb.panTro6.gapOverlap.txt
    # 6998 bases of 3050398082 (0.000%) in intersection

    # only six items:
    zcat *.bed.gz | cut -f1-5
# chr12   113245358       113246724       chr12:113245359-113246724       633
# chr21   18867677        18869111        chr21:18867678-18869111 667
# chr8    127452459       127454559       chr8:127452460-127454559        1000
# chrX    52589209        52609380        chrX:52589210-52609380  1000
# chrY    7097196 7097619 chrY:7097197-7097619    199

#############################################################################
# tandemDups (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/tandemDups
    cd /hive/data/genomes/panTro6/bed/tandemDups
    time (~/kent/src/hg/utils/automation/doTandemDup.pl \
  -twoBit=/hive/data/genomes/panTro6/panTro6.unmasked.2bit panTro6) \
	> do.log 2>&1 &
    # real    96m33.185s

    cat fb.panTro6.tandemDups.txt
    # 133769723 bases of 3050398082 (4.385%) in intersection

    bigBedInfo panTro6.tandemDups.bb | sed -e 's/^/#  /;'
#  version: 4
#  fieldCount: 13
#  hasHeaderExtension: yes
#  isCompressed: yes
#  isSwapped: 0
#  extraIndexCount: 0
#  itemCount: 2,570,600
#  primaryDataSize: 65,661,880
#  primaryIndexSize: 259,136
#  zoomLevels: 9
#  chromCount: 4134
#  basesCovered: 1,636,830,812
#  meanDepth (of bases covered): 12.911632
#  minDepth: 1.000000
#  maxDepth: 474.000000
#  std of depth: 28.418142

#############################################################################
# run up idKeys files for chromAlias/ncbiRefSeq (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/idKeys
    cd /hive/data/genomes/panTro6/bed/idKeys

    time (doIdKeys.pl \
	-twoBit=/hive/data/genomes/panTro6/panTro6.unmasked.2bit \
	-buildDir=`pwd` panTro6) > do.log 2>&1 &
    # real    3m56.622s

    cat panTro6.keySignature.txt
    #  d56a60e851e19f2c5b842c27bb43b990

#############################################################################
# ucscToINSDC and ucscToRefSeq table/track (DONE - 2018-03-24 - Hiram)
    # the sequence here is working for a 'refseq' assembly
    # beware of a chrM situation may be specific depending upon what is
    # available in the assembly

    mkdir /hive/data/genomes/panTro6/bed/ucscToINSDC
    cd /hive/data/genomes/panTro6/bed/ucscToINSDC

    join -t$'\t' ../idKeys/panTro6.idKeys.txt \
	../../refseq/idKeysGenbank/genbankPanTro6.idKeys.txt | cut -f2- \
          | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \
            | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \
               | sort -k1,1 -k2,2n > ucscToINSDC.bed

    join -t$'\t' ../idKeys/panTro6.idKeys.txt \
	../../refseq/idKeys/refseqPanTro6.idKeys.txt | cut -f2- \
          | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \
            | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \
               | sort -k1,1 -k2,2n > ucscToRefSeq.bed

    # verify line counts are the same
    wc -l *.bed ../../chrom.sizes
    #	4344 ucscToINSDC.bed
    #	4346 ucscToRefSeq.bed
    #	4346 ../../chrom.sizes

    # need to find INSDC equivalents for chrY and chrM:
    egrep "chrY|chrM" *
ucscToRefSeq.bed:chrM   0       16554   NC_001643.1
ucscToRefSeq.bed:chrY   0       26350515        NC_006492.4

    # lookup those accessions at NCBI Entrez
    # NC_001643.1 = D38113.1
    # NC_006492.4 = DP000054.3
    printf "chrM\t0\t16554\tD38113.1\n" >> ucscToINSDC.bed
    printf "chrY\t0\t26350515\tNC_006492.4\n" >> ucscToINSDC.bed

    # resort the bed file:
    sort -k1,1 -k2,2n -o ucscToINSDC.bed ucscToINSDC.bed

    # verify chrM and chrY are correct:
    egrep "chrM|chrY" *
ucscToINSDC.bed:chrM    0       16554   D38113.1
ucscToINSDC.bed:chrY    0       26350515        NC_006492.4
ucscToRefSeq.bed:chrM   0       16554   NC_001643.1
ucscToRefSeq.bed:chrY   0       26350515        NC_006492.4

    # should be same line counts throughout:
    wc -l * ../../chrom.sizes
    #	4346 ucscToINSDC.bed
    #	4346 ucscToRefSeq.bed
    #	4346 ../../chrom.sizes

    export chrSize=`cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1`
    echo $chrSize
    # 27
    # use the $chrSize in this sed
    sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \
         | hgLoadSqlTab panTro6 ucscToINSDC stdin ucscToINSDC.bed
    # should be the same for ucscToRefSeq:
    export chrSize=`cut -f1 ucscToRefSeq.bed | awk '{print length($0)}' | sort -n | tail -1`
    echo $chrSize
    #  27
    sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \
       | sed -e 's/INSDC/RefSeq/g;' > ucscToRefSeq.sql
    hgLoadSqlTab panTro6 ucscToRefSeq ./ucscToRefSeq.sql ucscToRefSeq.bed

    # checkTableCoords should be silent
    checkTableCoords panTro6
    # each should cover %100 entirely:
    featureBits -countGaps panTro6 ucscToINSDC
    # 3050398082 bases of 3050398082 (100.000%) in intersection

    featureBits -countGaps panTro6 ucscToRefSeq
    # 3050398082 bases of 3050398082 (100.000%) in intersection

#########################################################################
# add chromAlias table (DONE - 2018-03-24 - Hiram)

    mkdir /hive/data/genomes/panTro6/bed/chromAlias
    cd /hive/data/genomes/panTro6/bed/chromAlias

    hgsql -N -e 'select chrom,name from ucscToRefSeq;' panTro6 \
        | sort -k1,1 > ucsc.refseq.tab
    hgsql -N -e 'select chrom,name from ucscToINSDC;' panTro6 \
        | sort -k1,1 > ucsc.genbank.tab

    ~/kent/src/hg/utils/automation/chromAlias.pl ucsc.*.tab \
	> panTro6.chromAlias.tab

for t in refseq genbank
do
  c0=`cat ucsc.$t.tab | wc -l`
  c1=`grep $t panTro6.chromAlias.tab | wc -l`
  ok="OK"
  if [ "$c0" -ne "$c1" ]; then
     ok="ERROR"
  fi
  printf "# checking $t: $c0 =? $c1 $ok\n"
done
# checking refseq: 4346 =? 4346 OK
# checking genbank: 4346 =? 4346 OK

    hgLoadSqlTab panTro6 chromAlias ~/kent/src/hg/lib/chromAlias.sql \
        panTro6.chromAlias.tab

#########################################################################
# fixup search rule for assembly track/gold table (DONE - 2018-03-24 - Hiram)
    cd ~/kent/src/hg/makeDb/trackDb/chimp/panTro6

    # preview prefixes and suffixes:
    hgsql -N -e "select frag from gold;" panTro6 \
      | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c | sed -e 's/^/#\t/;'
#            14 AACZ.1
#            21 AC.1
#            90 AC.2
#            53 AC.3
#            22 AC.4
#            11 AC.5
#             3 AC.6
#             2 BS.2
#          5037 NBAG.1
#             1 NC_.1
               
    # implies a rule: '[ABN][ACSB][ACGZ0-9_]+(\.[0-9]+)?'

    # verify this rule will find them all and eliminate them all:
    hgsql -N -e "select frag from gold;" panTro6 | wc -l
    # 5254

    hgsql -N -e "select frag from gold;" panTro6 \
       | egrep -e '[ABN][ACSB][ACGZ0-9_]+(\.[0-9]+)?' | wc -l
    # 5254

    hgsql -N -e "select frag from gold;" panTro6 \
       | egrep -v -e '[ABN][ACSB][ACGZ0-9_]+(\.[0-9]+)?' | wc -l
    # 0

    # hence, add to trackDb/chicken/panTro6/trackDb.ra
searchTable gold
shortCircuit 1
termRegex [ABN][ACSB][ACGZ0-9_]+(\.[0-9]+)?
query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%'
searchPriority 8

    # verify searches work in the position box

#############################################################################
# running repeat masker (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/repeatMasker
    cd /hive/data/genomes/panTro6/bed/repeatMasker
    time  (doRepeatMasker.pl -buildDir=`pwd` \
        -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \
        -smallClusterHub=ku panTro6) > do.log 2>&1 &
    # real    1035m8.587s

    cat faSize.rmsk.txt
    egrep "bases|Total|masked" faSize.rmsk.txt \
	| fold -w 75 -s  | sed -e 's/^/# /;'
# 3050398082 bases (31805092 N's 3018592990 real 1405499360 upper 1613093630 
# lower) in 4346 sequences in 1 files
# Total size: mean 701886.4 sd 9434013.2 min 223 (chrUn_NW_019937062v1) max 
# 224244399 (chr1) median 29187
# %52.88 masked total, %53.44 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 panTro6 rmsk
    # 1613093500 bases of 3050398082 (52.881%) in intersection
    # real    0m35.581s

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

    # faster way to get the same result on high contig count assemblies:
    time hgsql -N -e 'select genoName,genoStart,genoEnd from rmsk;' panTro6 \
        | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total"
    # total 1613093500.000000
    # real    0m38.395s

##########################################################################
# running simple repeat (DONE - 2018-03-24 - Hiram)

    mkdir /hive/data/genomes/panTro6/bed/simpleRepeat
    cd /hive/data/genomes/panTro6/bed/simpleRepeat
    # using trf409 6 here as similar size to genome (human == 6)
    time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \
        -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \
        -trf409 6 panTro6) > do.log 2>&1 &
    # real    273m49.147s

    cat fb.simpleRepeat
    # 222649611 bases of 3018592990 (7.376%) in intersection

    bigBedInfo *.bb | sed -e 's/^/# /;'
# version: 4
# fieldCount: 16
# hasHeaderExtension: yes
# isCompressed: yes
# isSwapped: 0
# extraIndexCount: 0
# itemCount: 996,217
# primaryDataSize: 31,637,595
# primaryIndexSize: 202,580
# zoomLevels: 10
# chromCount: 4274
# basesCovered: 222,649,611
# meanDepth (of bases covered): 1.833462
# minDepth: 1.000000
# maxDepth: 88.000000
# std of depth: 2.284835

    # adding this trfMask to the other masking
    cd /hive/data/genomes/panTro6

    # when using the Window Masker result:
#    twoBitMask bed/windowMasker/panTro6.cleanWMSdust.2bit \
#       -add bed/simpleRepeat/trfMask.bed  panTro6.2bit
    #   you can safely ignore the warning about fields >= 13

    # when using Rmsk results, add to rmsk after it is done:
    twoBitMask panTro6.rmsk.2bit \
        -add bed/simpleRepeat/trfMask.bed panTro6.2bit
    #   you can safely ignore the warning about fields >= 13

    twoBitToFa panTro6.2bit stdout | faSize stdin > faSize.panTro6.2bit.txt
    egrep "bases|Total|masked" faSize.panTro6.2bit.txt \
	| fold -w 75 -s  | sed -e 's/^/# /;'
# 3050398082 bases (31805092 N's 3018592990 real 1402090427 upper 1616502563 
# lower) in 4346 sequences in 1 files
# Total size: mean 701886.4 sd 9434013.2 min 223 (chrUn_NW_019937062v1) max 
# 224244399 (chr1) median 29187
# %52.99 masked total, %53.55 masked real

    # reset the symlink
    rm /gbdb/panTro6/panTro6.2bit
    ln -s `pwd`/panTro6.2bit /gbdb/panTro6/panTro6.2bit

#########################################################################
# CREATE MICROSAT TRACK (DONE - 2018-03-24 - Hiram)
    ssh hgwdev
    mkdir /cluster/data/panTro6/bed/microsat
    cd /cluster/data/panTro6/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 panTro6 microsat microsat.bed
    # Read 28577 elements of size 4 from microsat.bed

##########################################################################
## WINDOWMASKER (DONE - 2018-03-24 - Hiram)

    mkdir /hive/data/genomes/panTro6/bed/windowMasker
    cd /hive/data/genomes/panTro6/bed/windowMasker
    time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \
        -dbHost=hgwdev panTro6) > do.log 2>&1
    # real    188m5.912s

    # Masking statistics
    cat faSize.panTro6.cleanWMSdust.txt
    egrep "bases|Total|masked" faSize.panTro6.cleanWMSdust.txt \
	| fold -w 75 -s  | sed -e 's/^/# /;'
# 3050398082 bases (31805092 N's 3018592990 real 1812556664 upper 1206036326 
# lower) in 4346 sequences in 1 files
# Total size: mean 701886.4 sd 9434013.2 min 223 (chrUn_NW_019937062v1) max 
# 224244399 (chr1) median 29187
# %39.54 masked total, %39.95 masked real

    cat fb.panTro6.rmsk.windowmaskerSdust.txt
    # 971547886 bases of 3050398082 (31.850%) in intersection

#############################################################################
# ncbiRefSeq (DONE - 2018-03-24 - Hiram)

    # can be run up after ucscToRefSeq table is constructed
    mkdir /hive/data/genomes/panTro6/bed/ncbiRefSeq
    cd /hive/data/genomes/panTro6/bed/ncbiRefSeq

    # adjust the name arguments
    time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \
      -bigClusterHub=ku -dbHost=hgwdev \
      -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \
      refseq vertebrate_mammalian Pan_troglodytes \
      GCF_002880755.1_Clint_PTRv2 panTro6) > do.log 2>&1
    # real    10m2.131s

    # started this too early.  It needs to run after the panTro6.2bit file
    # has been masked.  Rerunning from process step:
    time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \
      -continue=process -bigClusterHub=ku -dbHost=hgwdev \
      -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \
      refseq vertebrate_mammalian Pan_troglodytes \
      GCF_002880755.1_Clint_PTRv2 panTro6) > process.log 2>&1 &

    # real    8m52.047s

    cat fb.ncbiRefSeq.panTro6.txt
    #  81659062 bases of 2476453204 (3.297%) in intersection

#############################################################################
# cpgIslands - (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/cpgIslands
    cd /hive/data/genomes/panTro6/bed/cpgIslands
    time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \
      -workhorse=hgwdev -smallClusterHub=ku panTro6) > do.log 2>&1 &
    # real    5m35.611s

    cat fb.panTro6.cpgIslandExt.txt
    # 22143193 bases of 3018592990 (0.734%) in intersection

##############################################################################
# genscan - (DONE - 2018-03-24 - Hiram)
    mkdir /hive/data/genomes/panTro6/bed/genscan
    cd /hive/data/genomes/panTro6/bed/genscan
    time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \
      -bigClusterHub=ku panTro6) > do.log 2>&1 &
    # real    182m50.882s

    cat fb.panTro6.genscan.txt
    # 53595000 bases of 3018592990 (1.775%) in intersection

    cat fb.panTro6.genscanSubopt.txt
    # 51298399 bases of 3018592990 (1.699%) in intersection

    bigBedInfo panTro6.genscan.bb | sed -e 's/^/# /;'
# version: 4
# fieldCount: 12
# hasHeaderExtension: yes
# isCompressed: yes
# isSwapped: 0
# extraIndexCount: 0
# itemCount: 41,507
# primaryDataSize: 2,449,724
# primaryIndexSize: 38,244
# zoomLevels: 8
# chromCount: 871
# basesCovered: 1,998,870,176
# meanDepth (of bases covered): 1.000000
# minDepth: 1.000000
# maxDepth: 1.000000
# std of depth: 0.000000

#############################################################################
# augustus gene track (DONE - 2018-03-24 - Hiram)

    mkdir /hive/data/genomes/panTro6/bed/augustus
    cd /hive/data/genomes/panTro6/bed/augustus
    time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \
     -species=human -dbHost=hgwdev -workhorse=hgwdev panTro6) > do.log 2>&1 &
    # real    98m3.215s

    cat fb.panTro6.augustusGene.txt
    # 50421577 bases of 3018592990 (1.670%) in intersection

     bigBedInfo  panTro6.augustus.bb | sed -e 's/^/# /;'
# version: 4
# fieldCount: 20
# hasHeaderExtension: yes
# isCompressed: yes
# isSwapped: 0
# extraIndexCount: 0
# itemCount: 30,411
# primaryDataSize: 2,164,577
# primaryIndexSize: 28,960
# zoomLevels: 7
# chromCount: 495
# basesCovered: 1,295,044,049
# meanDepth (of bases covered): 1.295008
# minDepth: 1.000000
# maxDepth: 4.000000
# std of depth: 0.662334

#############################################################################
# lastz/chain/net swap human/hg38 (DONE - 2018-03-24 - Hiram)
    # original alignment
    cd /hive/data/genomes/hg38/bed/lastzPanTro6.2018-03-24

    cat fb.hg38.chainPanTro6Link.txt
    # 2910625246 bases of 3049335806 (95.451%) in intersection
    cat fb.hg38.chainSynPanTro6Link.txt
    # 2890994068 bases of 3049335806 (94.807%) in intersection
    cat fb.hg38.chainRBestPanTro6Link.txt
    # 2743876160 bases of 3049335806 (89.983%) in intersection

    # running the swap
    mkdir /hive/data/genomes/panTro6/bed/blastz.hg38.swap
    cd /hive/data/genomes/panTro6/bed/blastz.hg38.swap
    time (doBlastzChainNet.pl -verbose=2 \
        -swap /hive/data/genomes/hg38/bed/lastzPanTro6.2018-03-24/DEF \
        -chainMinScore=5000 -chainLinearGap=medium \
        -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \
        -syntenicNet) > swap.log 2>&1
    # real    101m29.782s

    cat fb.panTro6.chainHg38Link.txt
    # 2814507414 bases of 3018592990 (93.239%) in intersection
    cat fb.panTro6.chainSynHg38Link.txt
    # 2800642800 bases of 3018592990 (92.780%) in intersection

    time (doRecipBest.pl -load -workhorse=hgwdev -buildDir=`pwd` \
        panTro6 hg38) > rbest.log 2>&1 &
    # real    102m47.533s

    cat fb.panTro6.chainRBestHg38Link.txt 
    # 2748571185 bases of 3018592990 (91.055%) in intersection

#############################################################################
# lastz/chain/net swap mouse/mm10 (DONE - 2018-03-24 - Hiram)

    # alignment to mouse/mm10:
    cd /hive/data/genomes/mm10/bed/lastzPanTro6.2018-03-24/

    cat fb.mm10.chainPanTro6Link.txt
    # 935720585 bases of 2652783500 (35.273%) in intersection

    cat fb.mm10.chainRBestPanTro6Link.txt
    # 891553355 bases of 2652783500 (33.608%) in intersection

    # and for the swap:
    mkdir /hive/data/genomes/panTro6/bed/blastz.mm10.swap
    cd /hive/data/genomes/panTro6/bed/blastz.mm10.swap

    time (doBlastzChainNet.pl -verbose=2 \
      /hive/data/genomes/mm10/bed/lastzPanTro6.2018-03-24/DEF \
        -swap -chainMinScore=3000 -chainLinearGap=medium \
          -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \
            -syntenicNet) > swap.log 2>&1
    #  real    78m57.631s

    cat fb.panTro6.chainMm10Link.txt
    # 934668641 bases of 3018592990 (30.964%) in intersection
    cat fb.panTro6.chainSynMm10Link.txt
    # 889944141 bases of 3018592990 (29.482%) in intersection

    time (doRecipBest.pl -load -workhorse=hgwdev -buildDir=`pwd` \
	panTro6 mm10) > rbest.log 2>&1 &
    # real    504m47.811s

    cat fb.panTro6.chainRBestMm10Link.txt
    # 890065520 bases of 3018592990 (29.486%) in intersection

##############################################################################
# Create kluster run files (DONE - 2018-03-24 - Hiram)

    cd /hive/data/genomes/panTro6
    # numerator is panTro6 gapless bases "real" as reported by:
    featureBits -noRandom -noHap panTro6 gap
    # 28994139 bases of 2803637675 (1.034%) in intersection
    #                   ^^^

    # 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 \( 2803637675 / 2861349177 \) \* 1024
    #  ( 2803637675 / 2861349177 ) * 1024 = 1003.346604

    # ==> use -repMatch=1000 same as was panTro5
    cd /hive/data/genomes/panTro6
    blat panTro6.2bit \
         /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/panTro6.11.ooc \
        -repMatch=1000
    #  Wrote 36740 overused 11-mers to jkStuff/panTro6.11.ooc
    # panTro5 at repMatch=1000 was:
    #  Wrote 38245 overused 11-mers to jkStuff/panTro5.11.ooc

    #   check non-bridged gaps to see what the typical size is:
    hgsql -N \
        -e 'select * from gap where bridge="no" order by size;' panTro6 \
        | sort -k7,7nr | ave -col=7 stdin | sed -e 's/^/# /;'
    # there are many at 100 bases:
# Q1 100.000000
# median 100.000000
# Q3 100.000000
# average 10930.290698
# min 10.000000
# max 500001.000000
# count 86
# total 940005.000000
# standard deviation 56419.602180

    # profile:
    hgsql -N -e 'select * from gap where bridge="no" order by size;' panTro6 \
	| cut -f7 | sort | uniq -c
      1 10
     74 100
      1 10223
      1 131931
      1 15585
      1 29588
      1 32403
      1 33521
      1 500001
      2 50001
      1 70431
      1 8910

    # therefore, minimum gap size of 100, this is doing non-bridged gaps only
  gapToLift -verbose=2 -minGap=100 panTro6 jkStuff/panTro6.nonBridged.lft \
       -bedFile=stdout | sort -k1,1 -k2,2n > jkStuff/panTro6.nonBridged.bed

##############################################################################
# LIFTOVER TO panTro5 (DONE - 2018-03-24 - Hiram)
    ssh hgwdev
    mkdir /hive/data/genomes/panTro6/bed/blat.panTro5.2018-03-24
    cd /hive/data/genomes/panTro6/bed/blat.panTro5.2018-03-24
    time (doSameSpeciesLiftOver.pl -verbose=2 -buildDir=`pwd` \
	-ooc=/hive/data/genomes/panTro6/jkStuff/panTro6.11.ooc \
        -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \
         panTro6 panTro5) > do.log 2>&1
    # real    464m42.018s

    # verify the convert link on the test browser is now active from panTro6 to
    # panTro5

##############################################################################
# GENBANK AUTO UPDATE (DONE - 2018-03-25 - Hiram)
    ssh hgwdev
    cd $HOME/kent/src/hg/makeDb/genbank
    git pull
    # /cluster/data/genbank/data/organism.lst shows:
    # #organism      mrnaCnt   estCnt  refSeqCnt
    # Pan troglodytes 3548    5065    2048
    # Pan troglodytes schweinfurthii  7       0       0
    # Pan troglodytes troglodytes     42      0       0
    # Pan troglodytes verus   126     12315   0

    # edit etc/genbank.conf to add panTro6 just before panTro5
# panTro6 (chimp - Pan troglodytes - refseq GCF_002880755.1 - taxId 9598)
panTro6.serverGenome = /hive/data/genomes/panTro6/panTro6.2bit
panTro6.clusterGenome = /hive/data/genomes/panTro6/panTro6.2bit
panTro6.ooc = /hive/data/genomes/panTro6/jkStuff/panTro6.11.ooc
panTro6.lift = /hive/data/genomes/panTro6/jkStuff/panTro6.nonBridged.lft
panTro6.perChromTables = no
panTro6.refseq.mrna.native.pslCDnaFilter  = ${finished.refseq.mrna.native.pslCDnaFilter}
panTro6.refseq.mrna.xeno.pslCDnaFilter    = ${finished.refseq.mrna.xeno.pslCDnaFilter}
panTro6.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter}
panTro6.genbank.mrna.xeno.pslCDnaFilter   = ${finished.genbank.mrna.xeno.pslCDnaFilter}
panTro6.genbank.est.native.pslCDnaFilter  = ${finished.genbank.est.native.pslCDnaFilter}
panTro6.genbank.est.xeno.pslCDnaFilter    = ${finished.genbank.est.xeno.pslCDnaFilter}
panTro6.downloadDir = panTro6
# defaults yes: genbank.mrna.native.load genbank.mrna.native.loadDesc
# yes: genbank.est.native.load refseq.mrna.native.load
# yes: refseq.mrna.native.loadDesc refseq.mrna.xeno.load
# yes: refseq.mrna.xeno.loadDesc
# defaults no: genbank.mrna.xeno.load genbank.mrna.xeno.loadDesc
# no: genbank.est.native.loadDesc genbank.est.xeno.load
# no: genbank.est.xeno.loadDesc
# DO NOT NEED genbank.mrna.xeno except for human, mouse
# have been doing this for chimp too:
panTro6.genbank.mrna.xeno.load = yes

    # verify stated file paths do exist:
    grep panTro6 etc/genbank.conf | egrep "Genome|ooc|lift" \
	| awk '{print $NF}' | xargs ls -og
-rw-rw-r-- 1    146968 Mar 24 20:30 /hive/data/genomes/panTro6/jkStuff/panTro6.11.ooc
-rw-rw-r-- 1    258184 Mar 25 14:25 /hive/data/genomes/panTro6/jkStuff/panTro6.nonBridged.lft
-rw-rw-r-- 1 795113145 Mar 24 20:22 /hive/data/genomes/panTro6/panTro6.2bit
-rw-rw-r-- 1 795113145 Mar 24 20:22 /hive/data/genomes/panTro6/panTro6.2bit

    # add panTro6 to:
    #   etc/align.dbs etc/hgwdev.dbs
    git commit -m 'adding panTro6/cat refs #21043' \
	etc/genbank.conf etc/align.dbs etc/hgwdev.dbs
    git push
    # update /cluster/data/genbank/:
    make etc-update

    # XXX a few days later the genbank tables will be in the database

#############################################################################
#  BLATSERVERS ENTRY (DONE - 2018-08-26 - 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 ("panTro6", "blat1c", "17904", "1", "0"); \
	INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
	VALUES ("panTro6", "blat1c", "17905", "0", "1");' \
	    hgcentraltest
    #	test it with some sequence

##############################################################################
## reset default position to same as what panTro5 has: FOXP2
##  (DONE - 2018-04-09 - Hiram)

    ssh hgwdev
    hgsql -e 'update dbDb set defaultPos="chr7:111311998-111585728"
	where name="panTro6";' hgcentraltest

##############################################################################
# all.joiner update, downloads and in pushQ - (DONE - 2018-04-09 - Hiram)
    cd $HOME/kent/src/hg/makeDb/schema
    ~/kent/src/hg/utils/automation/verifyBrowser.pl panTro6
# 66 tables in database panTro6 - Chimp, Pan troglodytes
# verified 66 tables in database panTro6, 0 extra tables, 24 optional tables
# NCBI RefSeq genes     10 optional tables
# chainNetRBestHg38     3 optional tables
# chainNetRBestMm10     3 optional tables
# chainNetSynHg38       3 optional tables
# chainNetSynMm10       3 optional tables
# gapOverlap    1 optional tables
# tandemDups    1 optional tables
# 13 genbank tables found
# verified 29 required tables, 0 missing tables
# hg38 chainNet to panTro6 found 3 required tables
# mm10 chainNet to panTro6 found 3 required tables
# hg38 chainNet RBest and syntenic to panTro6 found 6 optional tables
# mm10 chainNet RBest and syntenic to panTro6 found 3 optional tables
# liftOver to previous versions: 1, from previous versions: 1

    # fixup all.joiner until this is a clean output
    joinerCheck -database=panTro6 -tableCoverage all.joiner
    joinerCheck -database=panTro6 -times all.joiner
    joinerCheck -database=panTro6 -keys all.joiner

    cd /hive/data/genomes/panTro6
    # clean up obsolete trackDb work, assuming you have already checked in
    # these trackDb files into the source tree
    rm -fr TemporaryTrackDbCheckout

    time (makeDownloads.pl -workhorse=hgwdev panTro6) > downloads.log 2>&1
    #  real    25m49.888s

    #   now ready for pushQ entry
    mkdir /hive/data/genomes/panTro6/pushQ
    cd /hive/data/genomes/panTro6/pushQ
  time (makePushQSql.pl -redmineList panTro6) > panTro6.pushQ.sql 2> stderr.out
    # remove the tandemDups and gapOverlap from the file list:
    sed -i -e "/tandemDups/d" redmine.panTro6.table.list
    sed -i -e "/Tandem Dups/d" redmine.panTro6.releaseLog.txt
    sed -i -e "/gapOverlap/d" redmine.panTro6.table.list
    sed -i -e "/Gap Overlaps/d" redmine.panTro6.releaseLog.txt

    #   check for errors in stderr.out, some are OK, e.g.:
    # WARNING: panTro6 does not have seq
    # WARNING: panTro6 does not have extFile

    # add the path names to the listing files in the redmine issue
    # in the three appropriate entry boxes:

/hive/data/genomes/panTro6/pushQ/redmine.panTro6.file.list
/hive/data/genomes/panTro6/pushQ/redmine.panTro6.releaseLog.txt
/hive/data/genomes/panTro6/pushQ/redmine.panTro6.table.list

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