Hello Peter, For all tracks, on the track description page near the top is a "Last Updated" date. This is the primary clue as the when the data was published to the browser. This does not mean that this is the freeze data from the source.
To find out about that, read the Source/Methods. You will find that some are more specific than others. If the data came from an external source, you may try contacting them to find out any particulars that are not included in our track description. Track contents should be interpreted as follows for these basic track types: 1) Mapping and Sequencing: Many will be based on the reference genome. Other were created by internal or external groups. 2) ENCODE: the freeze date is noted. 3) Genbank (RefSeq, EST, mRNA): Updated (added to) with the genbank incremental release data. 4) UCSC Genes and other Annotation tracks: Updated periodically or only once per assembly. In general, data would be more likely to be updated for the most current version of a genome's assembly then it would be for an older version (ENCODE is an exception to this - as it is still primarily based on hg18, while hg19 is the most current NCBI release). Digging into the details: If you wish to, examining what we call the "makedoc" for any assembly may provide more details (when data was downloaded from a source, etc). Makedocs are in the kent source tree under: src/hg/makeDb/doc/ one per assembly. Hopefully this helps, Thanks, Jennifer --------------------------------- Jennifer Jackson UCSC Genome Informatics Group http://genome.ucsc.edu/ On 5/3/10 6:44 AM, Peter Frommolt wrote: > Dear UCSC team, > > I have a question regarding the releases of the Human Genome assembly. > It is almost commonplace that the NCBI36.1 build corresponds to hg18 and > GRCh37 corresponds to hg19. At the NBCI, however, there also exist > annotation releases 36.2, 36.3 for which I'm not sure whether there is > any equivalent in the UCSC databases. Is there no tracking of version > numbers for the _annotation_ of the Human Genome at the UCSC? I mean, > version control is an important issue for reproducibility of results. > > Best regards, > _______________________________________________ Genome maillist - [email protected] https://lists.soe.ucsc.edu/mailman/listinfo/genome
