0d886f89f8374347018e9c0620adf111d4e7078e
hiram
  Thu Jan 4 12:22:56 2024 -0800
now working with dynamic blat on genome sizes over 4 Gb no redmine

diff --git src/hg/makeDb/doc/asmHubs/mkGenomes.pl src/hg/makeDb/doc/asmHubs/mkGenomes.pl
index 966fa59..5b5e869 100755
--- src/hg/makeDb/doc/asmHubs/mkGenomes.pl
+++ src/hg/makeDb/doc/asmHubs/mkGenomes.pl
@@ -18,32 +18,32 @@
   printf STDERR "result will write a local asmId.genomes.txt file for each hub\n";
   printf STDERR "and a local asmId.hub.txt file for each hub\n";
   printf STDERR "and a local asmId.groups.txt file for each hub\n";
   printf STDERR "and the output to stdout will be the overall genomes.txt\n";
   printf STDERR "index file for all genomes in the given list\n";
   exit 255;
 }
 
 my $downloadHost = "hgwdev";
 my @blatHosts = qw( dynablat-01 dynablat-01 );
 my @blatPorts = qw( 4040 4040 );
 my $blatHostDomain = ".soe.ucsc.edu";
 my $groupsTxt = `cat ~/kent/src/hg/makeDb/doc/asmHubs/groups.txt`;
 
 ################### writing out hub.txt file, twice ##########################
-sub singleFileHub($$$$$$$$$$$$$) {
-  my ($fh1, $fh2, $accessionId, $orgName, $descr, $asmId, $asmDate, $defPos, $taxId, $trackDb, $accessionDir, $buildDir, $chromAuthority) = @_;
+sub singleFileHub($$$$$$$$$$$$$$) {
+  my ($fh1, $fh2, $accessionId, $orgName, $descr, $asmId, $asmDate, $defPos, $taxId, $trackDb, $accessionDir, $buildDir, $chromAuthority, $hugeGenome) = @_;
   my @fhN;
   push @fhN, $fh1;
   push @fhN, $fh2;
 
   my %liftOverChain;	# key is 'otherDb' name, value is bbi path
   my %liftOverGz;	# key is 'otherDb' name, value is lift.over.gz file path
   my $hasChainNets = `ls -d $buildDir/trackData/lastz.* 2> /dev/null | wc -l`;
   chomp $hasChainNets;
   if ($hasChainNets) {
     printf STDERR "# hasChainNets: %d\t%s\n", $hasChainNets, $asmId;
     open (CH, "ls -d $buildDir/trackData/lastz.*|") or die "can not ls -d $buildDir/trackData/lastz.*";
     while (my $line = <CH>) {
       chomp $line;
       my $otherDb = basename($line);
       $otherDb =~ s/lastz.//;
@@ -86,44 +86,44 @@
     printf $fh "chromSizes %s.chrom.sizes.txt\n", $accessionId;
     if ( -s "${buildDir}/${asmId}.chromAlias.bb" ) {
       printf $fh "chromAliasBb %s.chromAlias.bb\n", $accessionId;
     } else {
       printf $fh "chromAlias %s.chromAlias.txt\n", $accessionId;
     }
     if ($chromAuthority =~ m/^chromAuthority/) {
        printf $fh "%s\n", $chromAuthority;
     }
     printf $fh "organism %s %s\n", $assemblyName, $asmDate;
     printf $fh "defaultPos %s\n", $defPos;
     printf $fh "scientificName %s\n", $descr;
     printf $fh "htmlPath html/%s.description.html\n", $asmId;
     # until blat server host is ready for hgdownload, avoid these lines
     if ($blatHosts[$fileCount] ne $downloadHost) {
-      printf $fh "blat %s%s %s dynamic $accessionDir/$accessionId\n", $blatHosts[$fileCount], $blatHostDomain, $blatPorts[$fileCount];
-      printf $fh "transBlat %s%s %s dynamic $accessionDir/$accessionId\n", $blatHosts[$fileCount], $blatHostDomain, $blatPorts[$fileCount];
-      printf $fh "isPcr %s%s %s dynamic $accessionDir/$accessionId\n", $blatHosts[$fileCount], $blatHostDomain, $blatPorts[$fileCount];
+      printf $fh "blat %s%s %s dynamic $accessionDir/$accessionId\n", $blatHosts[$fileCount], $blatHostDomain, $blatPorts[$fileCount]+$hugeGenome;
+      printf $fh "transBlat %s%s %s dynamic $accessionDir/$accessionId\n", $blatHosts[$fileCount], $blatHostDomain, $blatPorts[$fileCount]+$hugeGenome;
+      printf $fh "isPcr %s%s %s dynamic $accessionDir/$accessionId\n", $blatHosts[$fileCount], $blatHostDomain, $blatPorts[$fileCount]+$hugeGenome;
     }
     foreach my $otherDb (sort keys %liftOverGz) {
        printf $fh "liftOver.%s %s\n", $otherDb, $liftOverGz{$otherDb};
     }
     printf $fh "\n";
     foreach my $tdbLine (@tdbLines) {
       printf $fh "%s\n", $tdbLine;
     }
     ++$fileCount;
   }
-}
+}	#	sub singleFileHub($$$$$$$$$$$$$$)
 
 ##############################################################################
 my $home = $ENV{'HOME'};
 my $toolsDir = "$home/kent/src/hg/makeDb/doc/asmHubs";
 
 my $blatHost = shift;
 my $blatPort = shift;
 my $inputList = shift;
 my $orderList = $inputList;
 if ( ! -s "$orderList" ) {
   $orderList = $toolsDir/$inputList;
 }
 
 my @orderList;	# asmId of the assemblies in order from the *.list files
 # the order to read the different .list files:
@@ -225,104 +225,112 @@
   printf "twoBitPath ../%s/%s/%s.2bit\n", $accessionDir, $accessionId, $accessionId;
   printf "twoBitBptUrl ../%s/%s/%s.2bit.bpt\n", $accessionDir, $accessionId, $accessionId;
   printf "chromSizes ../%s/%s/%s.chrom.sizes.txt\n", $accessionDir, $accessionId, $accessionId;
 
   # wait until code gets out for v429 release before using chromAlias.bb
   # for the chromInfoPage display of hgTracks
   if ( -s "${buildDir}/${asmId}.chromAlias.bb" ) {
     printf "chromAliasBb ../%s/%s/%s.chromAlias.bb\n", $accessionDir, $accessionId, $accessionId;
   } else {
     printf "chromAlias ../%s/%s/%s.chromAlias.txt\n", $accessionDir, $accessionId, $accessionId;
   }
   if ($chromAuthority =~ m/^chromAuthority/) {
      printf "%s\n", $chromAuthority;
   }
   printf "organism %s %s\n", $assemblyName, $asmDate;
+  my $hugeGenome = 0;
+  my $fourGb = 2**32 - 1;
+  my $asmSize=`ave -col=2 $buildDir/$asmId.chrom.sizes | grep -w total | awk '{printf "%d", \$NF}'`;
+  chomp $asmSize;
+  $hugeGenome = 1 if ($asmSize > $fourGb);
   my $chrName=`head -1 $buildDir/$asmId.chrom.sizes | awk '{print \$1}'`;
   chomp $chrName;
   my $bigChrom=`head -1 $buildDir/$asmId.chrom.sizes | awk '{print \$NF}'`;
   chomp $bigChrom;
   my $oneThird = int($bigChrom/3);
   my $tenK = $oneThird + 10000;
   $tenK = $bigChrom if ($tenK > $bigChrom);
   my $defPos="${chrName}:${oneThird}-${tenK}";
   if ( -s "$buildDir/defaultPos.txt" ) {
     $defPos=`cat "$buildDir/defaultPos.txt"`;
     chomp $defPos;
   }
   printf "defaultPos %s\n", $defPos;
   printf "orderKey %d\n", $buildDone;
   printf "scientificName %s\n", $descr;
   printf "htmlPath ../%s/%s/html/%s.description.html\n", $accessionDir, $accessionId, $asmId;
   # until blat server host is ready for hgdownload, avoid these lines
   if ($blatHost ne $downloadHost) {
     if ( -s "${destDir}/$accessionId.trans.gfidx" ) {
-      printf "blat $blatHost$blatHostDomain $blatPort dynamic $accessionDir/$accessionId\n";
-    printf "transBlat $blatHost$blatHostDomain $blatPort dynamic $accessionDir/$accessionId\n";
-      printf "isPcr $blatHost$blatHostDomain $blatPort dynamic $accessionDir/$accessionId\n";
+      printf "blat $blatHost$blatHostDomain %d dynamic $accessionDir/$accessionId\n", $blatPort + $hugeGenome;
+    printf "transBlat $blatHost$blatHostDomain %d dynamic $accessionDir/$accessionId\n", $blatPort + $hugeGenome;
+      printf "isPcr $blatHost$blatHostDomain %d dynamic $accessionDir/$accessionId\n", $blatPort + $hugeGenome;
+    } else {
+      printf STDERR "# missing ${destDir}/$accessionId.trans.gfidx\n";
     }
   }
   printf "\n";
 
   # the original multi-file system:
   my $localHubTxt = "$buildDir/${asmId}.hub.txt";
   open (HT, ">$localHubTxt") or die "can not write to $localHubTxt";
   printf HT "hub %s genome assembly\n", $accessionId;
   printf HT "shortLabel %s\n", $orgName;
   printf HT "longLabel %s/%s/%s genome assembly\n", $orgName, $descr, $asmId;
   printf HT "genomesFile genomes.txt\n";
   printf HT "email hclawson\@ucsc.edu\n";
   printf HT "descriptionUrl html/%s.description.html\n", $asmId;
   close (HT);
 
   # try creating single file hub.txt, one for hgwdev, one for hgdownload
   my $downloadHubTxt = "$buildDir/${asmId}.download.hub.txt";
   open (DL, ">$downloadHubTxt") or die "can not write to $downloadHubTxt";
   $localHubTxt = "$buildDir/${asmId}.singleFile.hub.txt";
   open (HT, ">$localHubTxt") or die "can not write to $localHubTxt";
 
   singleFileHub(\*HT, \*DL, $accessionId, $orgName, $descr, $asmId, $asmDate,
-	$defPos, $taxId, $trackDb, $accessionDir, $buildDir, $chromAuthority);
+	$defPos, $taxId, $trackDb, $accessionDir, $buildDir, $chromAuthority,
+           $hugeGenome);
 
   my $localGenomesFile = "$buildDir/${asmId}.genomes.txt";
   open (GF, ">$localGenomesFile") or die "can not write to $localGenomesFile";
   printf GF "genome %s\n", $accessionId;
   printf GF "taxId %s\n", $taxId if (length($taxId) > 1);
   printf GF "trackDb trackDb.txt\n";
   printf GF "groups groups.txt\n";
   printf GF "description %s\n", $orgName;
   printf GF "twoBitPath %s.2bit\n", $accessionId;
   printf GF "twoBitBptUrl %s.2bit.bpt\n", $accessionId;
   printf GF "chromSizes %s.chrom.sizes.txt\n", $accessionId;
   if ( -s "${buildDir}/${asmId}.chromAlias.bb" ) {
     printf GF "chromAliasBb %s.chromAlias.bb\n", $accessionId;
   } else {
     printf GF "chromAlias %s.chromAlias.txt\n", $accessionId;
   }
   if ($chromAuthority =~ m/^chromAuthority/) {
      printf GF "%s\n", $chromAuthority;
   }
   printf GF "organism %s %s\n", $assemblyName, $asmDate;
   printf GF "defaultPos %s\n", $defPos;
   printf GF "scientificName %s\n", $descr;
   printf GF "htmlPath html/%s.description.html\n", $asmId;
   # until blat server host is ready for hgdownload, avoid these lines
   if ($blatHost ne $downloadHost) {
     if ( -s "${destDir}/$accessionId.trans.gfidx" ) {
-      printf GF "blat $blatHost$blatHostDomain $blatPort dynamic $accessionDir/$accessionId\n";
-      printf GF "transBlat $blatHost$blatHostDomain $blatPort dynamic $accessionDir/$accessionId\n";
-     printf GF "isPcr $blatHost$blatHostDomain $blatPort dynamic $accessionDir/$accessionId\n";
+      printf GF "blat $blatHost$blatHostDomain %d dynamic $accessionDir/$accessionId\n", $blatPort + $hugeGenome;
+      printf GF "transBlat $blatHost$blatHostDomain %d dynamic $accessionDir/$accessionId\n", $blatPort + $hugeGenome;
+     printf GF "isPcr $blatHost$blatHostDomain %d dynamic $accessionDir/$accessionId\n", $blatPort + $hugeGenome;
     }
   }
   close (GF);
 
   my $localGroups = "$buildDir/${asmId}.groups.txt";
   open (GR, ">$localGroups") or die "can not write to $localGroups";
   printf GR "%s", $groupsTxt;
   close (GR);
 }
 
 __END__
 
 description Mastacembelus armatus
 twoBitPath GCA_900324485.2_fMasArm1.2/trackData/addMask/GCA_900324485.2_fMasArm1.2.masked.2bit
 organism Zig-Zag eel