Post-Release-Checklist

Post Release Checklist for UCSC Genes

This page capture the things to remember after the release of UCSC Genes for the new assembly.

• Update the metadata in hgcentral:

set hgNearOk and hgPbOk = 1 in the dbDb table.

change defaultDb to the new assembly.

add a record in the gdbPdb table that points the assembly to the recent protein database.

• Check to see if otherOrgs.ra file is updated for this assembly in:

kent/src/hg/hgGene/hgGeneData/$org/$db/otherOrgs.ra

• Add the link to the protein database /goldenPath/proteinDB/proteinsYYMMDD/database/ in:

downloads.html file

Also, include the new assembly in this file: /goldenPath/proteinDB/proteinsYYMMDD/database/README.txt (and edit it out of the previous file).

• Exon Primer Links

Send an email to Tim Strom (http://ihg2.helmholtz-muenchen.de/) so that he knows we are releasing a new UCSC Gene set. He will need to prepare his website for the new UCSC IDs.

• HGNC Links to us

Contact Michael Lush: hgnc at genenames dot org to let him know about the update. He will need to rebuild his ucsc2hgnc mappings table.

• Galaxy (may want to update and/or pre-load these data)

Contact Anton at Galaxy to let him know about the update.

• UniProtKB

Create a speical file for the folks at UniProt. See this file for a sample of what they are looking for: hgwdev:/usr/local/apache/htdocs/goldenPath/hg18/UCSCGenes/uniProtToUcscGenes.txt

You can use this script to create the file: makeUniProtFile.csh

After creating this file, push it to the hgdownload machine. Then send an email to Elisabeth Gasteiger (Elisabeth -dot- Gasteiger -at- isb -dash- sib -dot- ch) letting her know we are releasing a new UCSC Gene set. She will download the file and change their links to our web site.

• Exoniphy

Exoniphy is a companion track (in the Genes group). It is created by Adam Seipel's group at Cornell. Contact Adam to ask him to create this track for this UCSC Genes release.

• Alert Ensembl:

Please e-mail this information to Glenn Proctor <glenn@ebi.ac.uk> and cc Ewan Birney <birney@ebi.ac.uk>. They would like UCSC Gene IDs, and coordinates and/or sequences. One problem is that they handle randoms differently to us and display individual scaffolds instead of an artificially created random chromosome like we do. We might also send them the knownToEnsembl table, too.