144522754f3e836dd0303944ada14c38ee9f28e3 rhead Mon Mar 12 14:38:54 2012 -0700 Added a new special case for rn4 (it now uses RGD Genes instead of a UCSC Genes build) and got rid of the checking the gdbPdb table to determine whether an assembly is a UCSC Genes assembly. Added a comment to the usage that the script should be checked to see if it is using up-to-date tables. diff --git src/utils/qa/makeUniProtFile.csh src/utils/qa/makeUniProtFile.csh index 4ab2b80..ceb399b 100755 --- src/utils/qa/makeUniProtFile.csh +++ src/utils/qa/makeUniProtFile.csh @@ -10,60 +10,62 @@ # Gene details pages. # ######################################## onintr cleanup set db='' set org='' set hasKGs='' set num='' if ($#argv != 1 ) then echo echo " Use this script to create a mapping of UniProt IDs to Gene IDs. \ UniProt will pick it up from our download server and use it to \ - create links from their web site to our gene details pages." + create links from their web site to our gene details pages. For the \ + gene sets from outside sources, check this script first to be sure it is \ + using the most up-to-date tables." echo echo " usage: db" echo exit 1 else set db=$argv[1] endif # run only from hgwdev if ( "$HOST" != "hgwdev" ) then echo "\nERROR: you must run this script on hgwdev!\n" exit 1 endif - -# check to see if this assembly has UCSC Genes -set hasKGs=`hgsql -Ne "SELECT genomeDb FROM gdbPdb" \ - hgcentraltest | grep "$db"` - -# based on whether or not this assembly has UCSC Genes, make the mapping file -if ( '' != $hasKGs ) then +# human and mouse use an official UCSC Genes build +if ( $db =~ "hg*" || $db =~ "mm*" ) then # get the data from the two KG-related tables hgsql -Ne "SELECT displayId, kgId FROM kgProtAlias" \ $db > $db.rawDataForUniProt hgsql -Ne "SELECT spId, kgId FROM kgSpAlias WHERE spId != ''" \ $db >> $db.rawDataForUniProt else #non-UCSC Gene assembly # each of the non-KG assemblies are treated a little differently + + if ( $db =~ "rn*" ) then + hgsql -Ne "hgsql -Ne "SELECT value, name FROM rgdGene2ToUniProt" $db > $db.rawDataForUniProt + endif + if ( $db =~ "dm*" ) then hgsql -Ne "SELECT alias, a.name FROM flyBase2004Xref AS a, \ flyBaseToUniProt AS b WHERE a.name=b.name AND alias != 'n/a'" \ $db > $db.rawDataForUniProt endif if ( $db =~ "ce*" ) then hgsql -Ne "SELECT acc, name FROM sangerGene AS a, uniProt.gene AS b \ WHERE a.proteinID=b.val and acc != 'n/a'" $db > $db.rawDataForUniProt endif if ( $db =~ "danRer*" ) then echo " \nERROR: Although a file could be generated for danRer, uniProt has" echo " decided that they do not want to do that mapping...yet\n" exit 1