"""
# SQLite lookup needs copying to not modify the input file which would mess up a rerun with -resume
cat $lookup > db.sqlite
msslookup ms1quant --dbfile db.sqlite -i ${krfns.join(' ')} --spectra ${mzmls.join(' ')} --quanttype kronik --mztol 20.0 --mztoltype ppm --rttol 5.0 
msslookup isoquant --dbfile db.sqlite -i ${isofns.join(' ')} --spectra ${isosamples.collect{ x -> x + '.mzML' }.join(' ')}
"""

"""
# SQLite lookup needs copying to not modify the input file which would mess up a rerun with -resume
cat $lookup > db.sqlite
msslookup ms1quant --dbfile db.sqlite -i ${krfns.join(' ')} --spectra ${mzmls.join(' ')} --quanttype kronik --mztol 20.0 --mztoltype ppm --rttol 5.0 
"""

"""
sqlite3 $speclookup "SELECT mzmlfilename, COUNT(*) FROM mzml JOIN mzmlfiles USING(mzmlfile_id) JOIN biosets USING(set_id) GROUP BY mzmlfilename" > amount_spectra_files
"""

"""
#!/usr/bin/env python
platesets = [\"${splates.join('", "')}\"]
platescans = {p: 0 for p in platesets}
fileplates = {fn: p for fn, p in zip([\"${mzmlfiles.join('", "')}\"], platesets)}
with open('nr_spec_per_file') as fp:
    for line in fp:
        fn, scans = line.strip('\\n').split('|')
        platescans[fileplates[fn]] += int(scans)
with open('scans_per_plate', 'w') as fp:
    for plate, scans in platescans.items():
        fp.write('{}\\t{}\\n'.format(plate, scans))
"""

"""
tryprev.py $tdb
cat $tdb decoy_${tdb} > db.fa
"""

"""
msgf_plus -Xmx16G -d $db -s $x -o "${sample}.mzid" -thread 12 -mod $mods -tda 0 -t 10.0ppm -ti -1,2 -m 0 -inst ${msgfinstrument} -e 1 -protocol ${msgfprotocol} -ntt 2 -minLength 7 -maxLength 50 -minCharge 2 -maxCharge 6 -n 1 -addFeatures 1
msgf_plus -Xmx3500M edu.ucsd.msjava.ui.MzIDToTsv -i "${sample}.mzid" -o out.mzid.tsv
rm ${db.baseName.replaceFirst(/\.fasta/, "")}.c*
"""

"""
echo $samples
mkdir mzids
count=1;for sam in ${samples.join(' ')}; do ln -s `pwd`/mzid\$count mzids/\${sam}.mzid; echo mzids/\${sam}.mzid >> metafile; ((count++));done
msgf2pin -o percoin.xml -e trypsin -P "decoy_" metafile
percolator -j percoin.xml -X perco.xml -N 500000 --decoy-xml-output -y
"""

"""
perco_to_tsv.py -p $perco --plates ${platenames.join(' ')} --fractions ${fractions.join(' ')}
"""

"""
msspsmtable merge -i psms* -o psms.txt
msspsmtable conffilt -i psms.txt -o filtpsm --confidence-better lower --confidence-lvl 0.01 --confcolpattern 'PSM q-value'
msspsmtable conffilt -i filtpsm -o filtpep --confidence-better lower --confidence-lvl 0.01 --confcolpattern 'peptide q-value'
# SQLite lookup needs copying to not modify the input file which would mess up a rerun with -resume
cat lookup > $psmlookup
msslookup psms -i filtpep --dbfile $psmlookup ${params.onlypeptides ? '' : "--fasta ${td == 'target' ? tdb : "${ddb} --decoy"}"} ${params.martmap ? "--map ${martmap}" : ''}
msspsmtable specdata -i filtpep --dbfile $psmlookup -o prepsms.txt
${!params.noquant ? "msspsmtable quant -i prepsms.txt -o qpsms.txt --dbfile $psmlookup --precursor ${params.isobaric && td=='target' ? '--isobaric' : ''}" : 'mv prepsms.txt qpsms.txt'}
sed 's/\\#SpecFile/SpectraFile/' -i qpsms.txt
${!params.onlypeptides ? "msspsmtable genes -i qpsms.txt -o gpsms --dbfile $psmlookup" : ''}
${!params.onlypeptides ? "msslookup proteingroup -i qpsms.txt --dbfile $psmlookup" : ''}
${!params.onlypeptides ? "msspsmtable proteingroup -i gpsms -o ${params.hirief ? "pgpsms" : "$outpsms"} --dbfile $psmlookup" : 'mv qpsms.txt pgpsms'}
${params.hirief ? "peptide_pi_annotator.py -i $trainingpep -p pgpsms --o $outpsms --stripcolpattern Strip --pepcolpattern Peptide --fraccolpattern Fraction --strippatterns ${allstrips.join(' ')} --intercepts ${allstrips.collect() { params.strips[it].intercept}.join(' ')} --widths ${allstrips.collect() { params.strips[it].fr_width}.join(' ')} --ignoremods \'*\'" : ''}
msspsmtable split -i ${outpsms} --bioset
"""

"""
# Create peptide table from PSM table, picking best scoring unique peptides
msspeptable psm2pep -i psms -o peptides --scorecolpattern svm --spectracol 1 ${!params.noquant && params.isobaric && td == 'target' ? "--isobquantcolpattern plex" : "" } ${!params.noquant ? "--ms1quantcolpattern area" : ""}
# Move peptide sequence to first column
paste <( cut -f ${col} peptides) <( cut -f 1-${col-1},${col+1}-500 peptides) > peptide_table.txt
# Create empty protein/gene/gene-symbol tables with only the identified accessions, will be filled later
echo Protein accession |tee proteins genes symbols
tail -n+2 psms|cut -f ${accolmap.proteins}|grep -v '\\;'|grep -v "^\$"|sort|uniq >> proteins
tail -n+2 psms|cut -f ${accolmap.genes}|grep -v '\\;'|grep -v "^\$"|sort|uniq >> genes
tail -n+2 psms|cut -f ${accolmap.assoc}|grep -v '\\;'|grep -v "^\$"|sort|uniq >> symbols
# Do isobaric quantification if necessary
${normalize && td == 'target' ? "msspsmtable isoratio -i psms -o proteinratios --protcol ${accolmap.proteins} --targettable proteins --isobquantcolpattern plex --minint 0.1 --denompatterns ${setdenoms[setname].join(' ')}" : 'touch proteinratios'}
${isoquant ? "msspsmtable isoratio -i psms -o pepisoquant --targettable peptide_table.txt --protcol ${accolmap.peptides} --isobquantcolpattern plex --minint 0.1 --denompatterns ${setdenoms[setname].join(' ')} ${normalize ? '--normalize median --norm-ratios proteinratios' : ''} > normratiosused" : ''}
${isoquant ? "mv pepisoquant peptide_table.txt" : ''}
# Create linear modeled q-values of peptides (modeled svm scores vs q-values) for more protein-FDR precision.
msspeptable modelqvals -i peptide_table.txt -o ${setname}_linmod --scorecolpattern svm --fdrcolpattern '^q-value'
"""

"""
mssprottable ms1quant -i proteins -o protms1 --psmtable psms --protcol ${accolmap[acctype]}
msspsmtable isoratio -i psms -o proteintable --protcol ${accolmap[acctype]} --targettable protms1 --isobquantcolpattern plex --minint 0.1 --denompatterns ${setdenoms[setname].join(' ')} ${normalize && td == 'target' ? '--norm-ratios pratios --normalize median': ''}
mssprottable bestpeptide -i proteintable -o bestpeptides --peptable peplinmod --scorecolpattern ${acctype == 'proteins' ? '\'^q-value\'' : '\'linear model\''} --logscore --protcol ${accolmap[acctype] + 1}
"""

"""
${td == 'target' && !params.noquant ? "mssprottable ms1quant -i proteins -o proteintable --psmtable psms --protcol ${accolmap[acctype]}" : 'mv proteins proteintable'}
mssprottable bestpeptide -i proteintable -o bestpeptides --peptable peplinmod --scorecolpattern ${acctype == 'proteins' ? '\'^q-value\'' : '\'linear model\''} --logscore --protcol ${accolmap[acctype] + 1}
"""

"""
mssprottable pickedfdr --picktype fasta --targetfasta $tfasta --decoyfasta $dfasta ${params.fastadelim ? "--fastadelim \'${params.fastadelim}\' --genefield ${params.genefield}" : ''} -i tbestpep --decoyfn dbestpep -o ${setname}_protfdr
"""

"""
mssprottable ${acctype == 'proteins' ? 'protfdr' : 'pickedfdr --picktype result'} -i tbestpep --decoyfn dbestpep -o ${setname}_protfdr
"""

"""
# SQLite lookup needs copying to not modify the input file which would mess up a rerun with -resume
cat $lookup > db.sqlite
msslookup ${acctype == 'peptides' ? 'peptides --fdrcolpattern \'^q-value\' --peptidecol' : 'proteins --fdrcolpattern \'q-value\' --protcol'} 1 --dbfile db.sqlite -i ${tables.join(' ')} --setnames ${setnames.join(' ')} ${!params.noquant ? "--ms1quantcolpattern area" : ""}  ${!params.noquant && params.isobaric ? '--psmnrcolpattern quanted --isobquantcolpattern plex' : ''} ${acctype in ['genes', 'assoc'] ? "--genecentric ${acctype}" : ''}
${acctype == 'peptides' ? 'msspeptable build' : 'mssprottable build --mergecutoff 0.01'} --dbfile db.sqlite -o proteintable ${!params.noquant && params.isobaric ? '--isobaric' : ''} ${!params.noquant ? "--precursor": ""} --fdr ${acctype in ['genes', 'assoc'] ? "--genecentric ${acctype}" : ''} ${params.onlypeptides ? "--noncentric" : ''}
sed -i 's/\\#/Amount/g' proteintable
"""

"""
qc_psms.R ${setnames[0].size()} ${fractionation ? 'TRUE' : 'FALSE'} ${plates.join(' ')}
echo "<html><body>" > psmqc.html
for graph in psm-scans missing-tmt miscleav
  do
  [[ -e \$graph ]] && paste -d \\\\0  <(echo "<div class=\\"chunk\\" id=\\"\${graph}\\"><img src=\\"data:image/png;base64,") <(base64 -w 0 \$graph) <(echo '"></div>') >> psmqc.html
  done 
for graph in retentiontime precerror fryield msgfscore
  do
  for plateid in ${plates.join(' ')}
    do
    plate="PLATE___\${plateid}___\${graph}"
    [[ -e \$plate ]] && paste -d \\\\0  <(echo "<div class=\\"chunk \$plateid\\" id=\\"\${graph}\\"><img src=\\"data:image/png;base64,") <(base64 -w 0 \$plate) <(echo '"></div>') >> psmqc.html
    done 
  done
echo "</body></html>" >> psmqc.html
"""

"""
${normalize ? "count=1;for setn in ${setnames.join(' ')}; do echo '' >> norm\${count} ; tail -n+2 norm\${count} | sed \$'s/ - /\t'\${setn}\$'\t/'; ((count++)); done >> normtable" : ''}
qc_protein.R ${setnames.size()} ${acctype} $peptable ${normalize ? 'normtable' : ''}
echo "<html><body>" > featqc.html
for graph in featyield precursorarea coverage isobaric nrpsms nrpsmsoverlapping percentage_onepsm normfac ms1nrpeps;
  do
  [ -e \$graph ] && paste -d \\\\0  <(echo "<div class=\\"chunk\\" id=\\"\${graph}\\"><img src=\\"data:image/png;base64,") <(base64 -w 0 \$graph) <(echo '"></div>') >> featqc.html
  done 
echo "</body></html>" >> featqc.html
"""

"""
count=1; for ac in ${acctypes.join(' ')}; do mv feat\$count \$ac.html; ((count++)); done
qc_collect.py $params.name ${params.hirief ? "hirief" : "nofrac"} ${plates.join(' ')}
"""

"""
markdown_to_html.r $output_docs results_description.html
"""