SNP Track QA

If you have any general questions about snps this is a good resource: http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=helpsnpfaq

• updateTimes.csh
  • make sure have most recent version

• makedoc
  • located at src/hg/makeDb/doc/{build}.txt
  • make sure that exists in makedoc

• look at mysql tables
  • check that the number of exceptions listed in snp###Exceptions and snp###ExceptionDesc match
  • make sure that there are no exceptions in the snp###Exceptions that do not also exist in the snp###ExceptionDesc table
  • If possible, compare to old SNP tables for this species. Look for big jumps in rows numbers, especially when grouped by func type or exception type.
  • check that func types in snp### table are documented in the html, both in the methods and in the check boxes. If there are new func types, make sure they are displayed correctly in the browser
  • check that the weights column in snp### are only 1,2 or 3
  • look the names of the snps. Make sure that they make sense (no extraneous ones, all lower case, etc). The checkHgFindSpec tool with the -checkTermRegex option will check that all of the names in the snp### table match the regular expression specified in trackDb. There is not an option to run it on a single table, so just run it on the whole database and look for output specific to the snp table. For instance:

 checkHgFindSpec -checkTermRegex hg19

• check settings (hgTrackUi and clicking on individual snp)
  • make sure that all settings that you can select for are present in mysql tables and vice versa
  • make sure that methods mention all func types that you can select in trackUi
  • check that can turn different gene tracks on in the trackUi

• all details for one data point (click through everything)

• featureBits, featureBits gap (or runBits.chs which does both featureBits and featureBits gap), countPerChrom.csh, and checkCoverage.csh
  • look at featureBits/Coverage for main snp table as well as the dbCoding and the ortho table. Also check coverage of haplo chromosomes.
  • see how coverage compares to old snp track if possible. Also when open up the track, most snps should match to the same place (though some will change)

• index
  • click "show sizes" in PushQ

• runJoiner.csh

• checkTableCoords
  • look for illegal genome coordinates for all positional tables

• performance
  • make sure appropriate entries in hgFindSpec (this is very important for snps since the table is so large). Check that entries in hgFindSpec work by searching for some snps.
  • make sure track "turns off" when viewing large regions (table too large to load for large regions)
  • look at large snp insertions/deletions and check that display correctly in hgTracks and hgTrackUi
  • check that display looks good
  • positionalTblCheck - script that checks to see if the table is ordered for faster access

• description
  • check spelling and grammar
  • make sure external files mentioned at the end of the html page exist at ncbi
  • emails sanitized

• default position - is it turned on by default?

• table descriptions - view table schema

• release log and release log url

AT RELEASE BE SURE TO:

1. push tableDescriptions and hgFindSpec
2. edit trackDb.ra so that old snp track is hidden by default if needed
3. make announcement on genome announce
4. notify Mark D so he can update LS-SNP
5. notify Galt so he can default to most current snp track in genome graphs