c8828612e3aba9e6c3072768129de10abe337878 hiram Fri May 29 11:29:04 2020 -0700 now allow orderList to be specifically stated and begin common set of html functions refs #23734 diff --git src/hg/makeDb/doc/asmHubs/trackData.pl src/hg/makeDb/doc/asmHubs/trackData.pl index ab7922d..55b31ca 100755 --- src/hg/makeDb/doc/asmHubs/trackData.pl +++ src/hg/makeDb/doc/asmHubs/trackData.pl @@ -1,521 +1,502 @@ #!/usr/bin/env perl use strict; use warnings; +use FindBin qw($Bin); +use lib "$Bin"; +use commonHtml; use File::stat; my $argc = scalar(@ARGV); if ($argc < 3) { printf STDERR "usage: trackData.pl Name asmHubName [two column name list] > trackData.html\n"; printf STDERR "e.g.: trackData.pl Mammals mammals mammals.asmId.commonName.tsv > trackData.html\n"; printf STDERR "the name list is found in \$HOME/kent/src/hg/makeDb/doc/asmHubs/\n"; printf STDERR "\nthe two columns are 1: asmId (accessionId_assemblyName)\n"; printf STDERR "column 2: common name for species, columns separated by tab\n"; exit 255; } +my $home = $ENV{'HOME'}; +my $toolsDir = "$home/kent/src/hg/makeDb/doc/asmHubs"; + my $testOutput = 0; my $spliceOut = -1; if ($argc > 2) { for (my $i = 0; $i < $argc; ++$i) { if ($ARGV[$i] =~ /-test/) { $testOutput = 1; $spliceOut = $i; } } } if ($spliceOut != -1) { splice @ARGV, $spliceOut, 1; } my $Name = shift; my $asmHubName = shift; -my $commonNameOrder = shift; - -my $home = $ENV{'HOME'}; -my $toolsDir = "$home/kent/src/hg/makeDb/doc/asmHubs"; +my $inputList = shift; +my $orderList = $inputList; +if ( ! -s "$orderList" ) { + $orderList = $toolsDir/$inputList; +} -my @orderList; # asmId of the assemblies in order from the commonNameOrder file +my @orderList; # asmId of the assemblies in order from the orderList file my %commonName; # key is asmId, value is a common name, perhaps more appropriate # than found in assembly_report file # assembly_report my $vgpIndex = 0; $vgpIndex = 1 if ($Name =~ m/vgp/i); my $assemblyTotal = 0; # complete list of assemblies in this group my $asmCount = 0; # count of assemblies completed and in the table my $overallNucleotides = 0; my $overallSeqCount = 0; my $overallGapSize = 0; my $overallGapCount = 0; ############################################################################## # from Perl Cookbook Recipe 2.17, print out large numbers with comma delimiters: ############################################################################## sub commify($) { my $text = reverse $_[0]; $text =~ s/(\d\d\d)(?=\d)(?!\d*\.)/$1,/g; return scalar reverse $text } # ($itemCount, $percentCover) = oneTrackData($asmId, $track, $trackFile, $totalSize, $trackFb, $runDir); # might have a track feature bits file (trackFb), maybe not sub oneTrackData($$$$$$) { my ($asmId, $trackName, $file, $genomeSize, $trackFb, $runDir) = @_; # printf STDERR "# %s\n", $file; my $itemCount = 0; my $percentCover = 0; if (! -s "${file}") { if ($trackName eq "gapOverlap") { if (-s "${runDir}/$asmId.gapOverlap.bed.gz" ) { my $lineCount=`zcat "${runDir}/$asmId.gapOverlap.bed.gz" | head | wc -l`; chomp $lineCount; if (0 == $lineCount) { return("0", "0 %"); } else { return("n/a", "n/a"); } } } elsif ($trackName eq "gap") { return("0", "0 %"); } else { return("n/a", "n/a"); } } else { if ($file =~ m/.bw$/) { my $bigWigInfo = `bigWigInfo "$file" | egrep "basesCovered:|mean:" | awk '{print \$NF}' | xargs echo | sed -e 's/,//g;'`; chomp $bigWigInfo; my ($bases, $mean) = split('\s+', $bigWigInfo); $percentCover = sprintf("%.2f %%", 100.0 * $bases / $genomeSize); $itemCount = sprintf ("%.2f", $mean); # printf STDERR "# bigWigInfo %s %s %s\n", $itemCount, $percentCover, $file; } else { my $bigBedInfo = `bigBedInfo "$file" | egrep "itemCount:|basesCovered:" | awk '{print \$NF}' | xargs echo | sed -e 's/,//g;'`; chomp $bigBedInfo; my ($items, $bases) = split('\s', $bigBedInfo); $itemCount = commify($items); $percentCover = sprintf("%.2f %%", 100.0 * $bases / $genomeSize); # 56992654 bases of 2616369673 (2.178%) in intersection if ( -s "${trackFb}" ) { my ($itemBases, undef, undef, $noGapSize, undef) = split('\s+', `cat $trackFb`, 5); $percentCover = sprintf("%.2f %%", 100.0 * $itemBases / $noGapSize); } # printf STDERR "# bigBedInfo %s %s %s\n", $itemCount, $percentCover, $file; } } return ($itemCount, $percentCover); } ############################################################################## ### start the HTML output ############################################################################## sub startHtml() { my $timeStamp = `date "+%F"`; chomp $timeStamp; my $subSetMessage = "subset of $asmHubName only"; if ($asmHubName eq "vertebrate") { $subSetMessage = "subset of other ${asmHubName}s only"; } if ($vgpIndex) { + my $vgpSubset = "(set of primary assemblies)"; + if ($orderList =~ m/vgp.alternate/) { + $vgpSubset = "(set of alternate/haplotype assemblies)"; + } elsif ($orderList =~ m/vgp.trio/) { + $vgpSubset = "(set of trio assemblies, maternal/paternal)"; + } elsif ($orderList =~ m/vgp.legacy/) { + $vgpSubset = "(set of legacy/superseded assemblies)"; + } print <<"END" <!DOCTYPE HTML 4.01 Transitional> <!--#set var="TITLE" value="VGP - Vertebrate Genomes Project assembly hubs, track statistics" --> <!--#set var="ROOT" value="../.." --> <!--#include virtual="\$ROOT/inc/gbPageStartHardcoded.html" --> <h1>VGP - Vertebrate Genomes Project assembly hubs, track statistics</h1> <p> <a href='https://vertebrategenomesproject.org/' target=_blank> <img src='VGPlogo.png' width=280 alt='VGP logo'></a></p> <p> This assembly hub contains assemblies released by the <a href='https://vertebrategenomesproject.org/' target=_blank> -Vertebrate Genomes Project.</a> +Vertebrate Genomes Project.</a> $vgpSubset </p> END } else { print <<"END" <!DOCTYPE HTML 4.01 Transitional> <!--#set var="TITLE" value="$Name genomes assembly hubs, track statistics" --> <!--#set var="ROOT" value="../.." --> <!--#include virtual="\$ROOT/inc/gbPageStartHardcoded.html" --> <h1>$Name Genomes assembly hubs, track statistics</h1> <p> Assemblies from NCBI/Genbank/Refseq sources, $subSetMessage. </p> END } my $indexUrl = "index"; $indexUrl = "testIndex" if ($testOutput); my $asmStats = "asmStats"; $asmStats = "testAsmStats" if ($testOutput); print <<"END" <h3>See also: <a href='$indexUrl.html'>hub access</a>, <a href='$asmStats.html'>assembly statistics</a></h3><br> <h3>Data resource links</h3> NOTE: <em>Click on the column headers to sort the table by that column</em><br> The <em>link to genome browser</em> will attach only that single assembly to the genome browser.<br> The numbers are: item count (percent coverage)<br> Except for the gc5Base column which is: overall GC % average (percent coverage) END } # sub startHtml() # order of columns in the table # eliminated the ncbiGene track my @trackList = qw(ncbiRefSeq xenoRefGene augustus ensGene gc5Base allGaps assembly rmsk simpleRepeat windowMasker cpgIslandExtUnmasked); ############################################################################## ### start the table output ############################################################################## sub startTable() { # coordinate the order of these column headings with the @trackList listed above print '<table class="sortable" border="1"> <thead><tr><th>count</th> <th>common name<br>link to genome browser</th> <th class="sorttable_numeric">ncbiRefSeq</th> '; print " <th class=\"sorttable_numeric\">ncbiGene</th>\n" if ($testOutput); print ' <th class="sorttable_numeric">xenoRefGene</th> <th class="sorttable_numeric">augustus<br>genes</th> <th class="sorttable_numeric">Ensembl<br>genes</th> <th class="sorttable_numeric">gc5 base</th> '; if ($testOutput) { print " <th class=\"sorttable_numeric\">AGP<br>gap</th>\n"; print " <th class=\"sorttable_numeric\">all<br>gaps</th>\n"; } else { print " <th class=\"sorttable_numeric\">gaps</th>\n"; } print ' <th class="sorttable_numeric">assembly<br>sequences</th> <th class="sorttable_numeric">Repeat<br>Masker</th> <th class="sorttable_numeric">TRF<br>simpleRepeat</th> <th class="sorttable_numeric">window<br>Masker</th> '; if ($testOutput) { print ' <th class="sorttable_numeric">gap<br>Overlap</th> <th class="sorttable_numeric">tandem<br>Dups</th> <th class="sorttable_numeric">cpg<br>unmasked</th> <th class="sorttable_numeric">cpg<br>island</th> '; } else { print " <th class=\"sorttable_numeric\">cpg<br>islands</th>\n"; } print "</tr></thead><tbody>\n"; } # sub startTable() ############################################################################## ### end the table output ############################################################################## sub endTable() { my $commaNuc = commify($overallNucleotides); my $commaSeqCount = commify($overallSeqCount); my $commaGapSize = commify($overallGapSize); my $commaGapCount = commify($overallGapCount); my $percentDone = 100.0 * $asmCount / $assemblyTotal; my $doneMsg = ""; if ($asmCount < $assemblyTotal) { $doneMsg = sprintf(" (%d build completed, %.2f %% finished)", $asmCount, $percentDone); } my $columnCount = scalar(@trackList); my $colSpanFill = $columnCount - 1; if ($assemblyTotal > 1) { print <<"END" </tbody> <tfoot><tr><th>TOTALS:</th><td align=center colspan=$colSpanFill>total assembly count ${assemblyTotal}${doneMsg}</td> </tr></tfoot> </table> END } else { print <<"END" </tbody> </table> END } } # sub endTable() ############################################################################## ### end the HTML output ############################################################################## sub endHtml() { -# do not print these links for VGP index +&commonHtml::otherHubLinks($vgpIndex, $asmHubName); +&commonHtml::htmlFooter($vgpIndex, $asmHubName); -if ((0 == $vgpIndex) && ($asmHubName ne "viral")) { - printf "<p>\n<table border='1'><thead>\n<tr>"; - printf "<th>Assembly hubs index pages: </th>\n"; - printf "<th><a href='../primates/index.html'>Primates</a></th>\n"; - printf "<th><a href='../mammals/index.html'>Mammals</a></th>\n"; - printf "<th><a href='../birds/index.html'>Birds</a></th>\n"; - printf "<th><a href='../fish/index.html'>Fish</a></th>\n"; - printf "<th><a href='../vertebrate/index.html'>other vertebrates</a></th>\n"; - - printf "</tr><tr>\n"; - printf "<th>Hubs assembly statistics: </th>\n"; - printf "<th><a href='../primates/asmStats.html'>Primates</a></th>\n"; - printf "<th><a href='../mammals/asmStats.html'>Mammals</a></th>\n"; - printf "<th><a href='../birds/asmStats.html'>Birds</a></th>\n"; - printf "<th><a href='../fish/asmStats.html'>Fish</a></th>\n"; - printf "<th><a href='../vertebrate/asm.html'>other vertebrates</a></th>\n"; - - printf "</tr><tr>\n"; - printf "<th>Hubs track statistics: </th>\n"; - printf "<th><a href='../primates/trackData.html'>Primates</a></th>\n"; - printf "<th><a href='../mammals/trackData.html'>Mammals</a></th>\n"; - printf "<th><a href='../birds/trackData.html'>Birds</a></th>\n"; - printf "<th><a href='../fish/trackData.html'>Fish</a></th>\n"; - printf "<th><a href='../vertebrate/trackData.html'>other vertebrates</a></th>\n"; - - printf "</tr></thead>\n</table>\n</p>\n"; -} - -print <<"END" -</div><!-- closing gbsPage from gbPageStartHardcoded.html --> -</div><!-- closing container-fluid from gbPageStartHardcoded.html --> -<!--#include virtual="\$ROOT/inc/gbFooterHardcoded.html"--> -<script type="text/javascript" src="/js/sorttable.js"></script> -</body></html> -END } # sub endHtml() sub asmCounts($) { my ($chromSizes) = @_; my ($sequenceCount, $totalSize) = split('\s+', `ave -col=2 $chromSizes | egrep "^count|^total" | awk '{printf "%d\\n", \$NF}' | xargs echo`); return ($sequenceCount, $totalSize); } # my ($gapSize, $maskPerCent, $sizeNoGaps) = maskStats($faSizeTxt); sub maskStats($) { my ($faSizeFile) = @_; my $sizeNoGaps = `grep 'sequences in 1 file' $faSizeFile | awk '{print \$4}'`; my $gapSize = `grep 'sequences in 1 file' $faSizeFile | awk '{print \$3}'`; chomp $gapSize; $gapSize =~ s/\(//; my $totalBases = `grep 'sequences in 1 file' $faSizeFile | awk '{print \$1}'`; chomp $totalBases; my $maskedBases = `grep 'sequences in 1 file' $faSizeFile | awk '{print \$9}'`; chomp $maskedBases; my $maskPerCent = 100.0 * $maskedBases / $totalBases; return ($gapSize, $maskPerCent, $sizeNoGaps); } # grep "sequences in 1 file" GCA_900324465.2_fAnaTes1.2.faSize.txt # 555641398 bases (3606496 N's 552034902 real 433510637 upper 118524265 lower) in 50 sequences in 1 files sub gapStats($$) { my ($buildDir, $asmId) = @_; my $gapBed = "$buildDir/trackData/allGaps/$asmId.allGaps.bed.gz"; my $gapCount = 0; if ( -s "$gapBed" ) { $gapCount = `zcat $gapBed | awk '{print \$3-\$2}' | ave stdin | grep '^count' | awk '{print \$2}'`; } chomp $gapCount; return ($gapCount); } ############################################################################## ### tableContents() ############################################################################## sub tableContents() { my $asmCounted = 0; if ($testOutput) { # add extra columns during 'test' output # 0 1 2 3 4 5 6 # 7 8 9 10 11 12 13 # 14 # my @trackList = qw(ncbiRefSeq xenoRefGene augustus ensGene gc5Base gap allGaps assembly rmsk simpleRepeat windowMasker gapOverlap tandemDups cpgIslandExtUnmasked cpgIslandExt); # 0 1 2 3 4 5 6 # 7 8 9 10 # my @trackList = qw(ncbiRefSeq xenoRefGene augustus ensGene gc5Base allGaps assembly rmsk simpleRepeat windowMasker cpgIslandExtUnmasked); splice @trackList, 11, 0, "cpgIslandExt"; splice @trackList, 10, 0, "tandemDups"; splice @trackList, 10, 0, "gapOverlap"; splice @trackList, 5, 0, "gap"; splice @trackList, 1, 0, "ncbiGene"; } foreach my $asmId (@orderList) { my $tracksCounted = 0; my ($gcPrefix, $asmAcc, $asmName) = split('_', $asmId, 3); my $accessionId = sprintf("%s_%s", $gcPrefix, $asmAcc); my $accessionDir = substr($asmId, 0 ,3); $accessionDir .= "/" . substr($asmId, 4 ,3); $accessionDir .= "/" . substr($asmId, 7 ,3); $accessionDir .= "/" . substr($asmId, 10 ,3); my $buildDir = "/hive/data/genomes/asmHubs/refseqBuild/$accessionDir/$asmId"; if ($gcPrefix eq "GCA") { $buildDir = "/hive/data/genomes/asmHubs/genbankBuild/$accessionDir/$asmId"; } my $trackDb="$buildDir/${asmId}.trackDb.txt"; next if (! -s "$trackDb"); # assembly build not complete my $asmReport="$buildDir/download/${asmId}_assembly_report.txt"; if (! -s "$asmReport") { printf STDERR "# no assembly report:\n# %s\n", $asmReport; next; } my $chromSizes = "${buildDir}/${asmId}.chrom.sizes"; my $twoBit = "${buildDir}/trackData/addMask/${asmId}.masked.2bit"; if (! -s "$twoBit") { printf STDERR "# no 2bit file:\n# %s\n", $twoBit; printf "<tr><td align=right>%d</td>\n", ++$asmCount; printf "<td align=center>%s</td>\n", $accessionId; printf "<th colspan=15 align=center>missing masked 2bit file</th>\n"; printf "</tr>\n"; next; } my $faSizeTxt = "${buildDir}/${asmId}.faSize.txt"; if ( ! -s "$faSizeTxt" ) { printf STDERR "twoBitToFa $twoBit stdout | faSize stdin > $faSizeTxt\n"; print `twoBitToFa $twoBit stdout | faSize stdin > $faSizeTxt`; } my ($gapSize, $maskPerCent, $sizeNoGaps) = maskStats($faSizeTxt); $overallGapSize += $gapSize; my ($seqCount, $totalSize) = asmCounts($chromSizes); $overallSeqCount += $seqCount; # my $totalSize=`ave -col=2 $chromSizes | grep "^total" | awk '{printf "%d", \$NF}'`; $overallNucleotides += $totalSize; my $gapCount = gapStats($buildDir, $asmId); $overallGapCount += $gapCount; my $sciName = "notFound"; my $commonName = "notFound"; my $asmDate = "notFound"; my $itemsFound = 0; open (FH, "<$asmReport") or die "can not read $asmReport"; while (my $line = <FH>) { last if ($itemsFound > 5); chomp $line; $line =~ s/ //g;; $line =~ s/\s+$//g;; if ($line =~ m/Date:/) { if ($asmDate =~ m/notFound/) { ++$itemsFound; $asmDate = $line; $asmDate =~ s/.*:\s+//; } } elsif ($line =~ m/Organism name:/) { if ($sciName =~ m/notFound/) { ++$itemsFound; $commonName = $line; $sciName = $line; $commonName =~ s/.*\(//; $commonName =~ s/\)//; $commonName = $commonName{$asmId} if (exists($commonName{$asmId})); $sciName =~ s/.*:\s+//; $sciName =~ s/\s+\(.*//; } } } close (FH); my $hubUrl = "https://genome.ucsc.edu/h/$accessionId"; if ($testOutput) { $hubUrl = "https://genome-test.gi.ucsc.edu/h/$accessionId"; } printf "<tr><td align=right>%d</td>\n", ++$asmCount; printf "<td align=center><a href='%s' target=_blank>%s<br>%s</a></td>\n", $hubUrl, $commonName, $accessionId; foreach my $track (@trackList) { my $trackFile = "$buildDir/bbi/$asmId.$track"; my $trackFb = "$buildDir/trackData/$track/fb.$asmId.$track.txt"; my $runDir = "$buildDir/trackData/$track"; my ($itemCount, $percentCover); if ( "$track" eq "gc5Base" ) { $trackFile .= ".bw"; } else { $trackFile .= ".bb"; } my $customKey = ""; if ( "$track" eq "rmsk") { my $rmskStats = "$buildDir/trackData/repeatMasker/$asmId.rmsk.stats"; if (! -s "${rmskStats}") { my $faOut = "$buildDir/trackData/repeatMasker/$asmId.sorted.fa.out.gz"; if ( -s "$faOut") { my $items = `zcat "$faOut" | wc -l`; chomp $items; $itemCount = commify($items); my $masked = `grep masked "$buildDir/trackData/repeatMasker/faSize.rmsk.txt" | awk '{print \$4}' | sed -e 's/%//;'`; chomp $masked; $percentCover = sprintf("%.2f %%", $masked); open (RS, ">$rmskStats") or die "can now write to $rmskStats"; printf RS "%s\t%s\n", $itemCount, $percentCover; close (RS); } else { $itemCount = "n/a"; $percentCover = "n/a"; } } else { ($itemCount, $percentCover) = split('\s+', `cat $rmskStats`); chomp $percentCover; $customKey = sprintf("%.2f", $percentCover); $percentCover = sprintf("%.2f %%", $percentCover); } } else { # not the rmsk track ($itemCount, $percentCover) = oneTrackData($asmId, $track, $trackFile, $totalSize, $trackFb, $runDir); if (0 == $testOutput) { # if track ncbiRefSeq does not exist, try the ncbiGene track if ($trackDb eq "ncbiRefSeq" && $itemCount eq "n/a") { $runDir = "$buildDir/trackData/ncbiGene"; $trackFile = "$buildDir/bbi/$asmId.$track.bb"; ($itemCount, $percentCover) = oneTrackData($asmId, "ncbiGene", $trackFile, $totalSize, $trackFb, $runDir); } } if (($percentCover =~ m/%/) || ($percentCover !~ m#n/a#)) { $customKey = $percentCover; $customKey =~ s/[ %]+//; } } if (length($customKey)) { printf " <td align=right sorttable_customkey='%s'>%s<br>(%s)</td>\n", $customKey, $itemCount, $percentCover; } else { if ($itemCount eq "n/a") { printf " <td align=right>n/a</td>\n"; } else { printf " <td align=right>%s<br>(%s)</td>\n", $itemCount, $percentCover; } } $tracksCounted += 1 if ($itemCount ne "n/a"); } # foreach my $track (@trackList) printf "</tr>\n"; $asmCounted += 1; printf STDERR "# %03d\t%02d tracks\t%s\n", $asmCounted, $tracksCounted, $asmId; } } ############################################################################## ### main() ############################################################################## -open (FH, "<$toolsDir/${commonNameOrder}") or die "can not read ${commonNameOrder}"; +open (FH, "<${orderList}") or die "can not read ${orderList}"; while (my $line = <FH>) { next if ($line =~ m/^#/); chomp $line; my ($asmId, $commonName) = split('\t', $line); push @orderList, $asmId; $commonName{$asmId} = $commonName; ++$assemblyTotal; } close (FH); startHtml(); startTable(); tableContents(); endTable(); endHtml();