[BioC] SRAdb: is the database missing some entries? (Ben Woodcroft)

Jack Zhu zhujack at mail.nih.gov
Wed Jun 20 23:30:22 CEST 2012


Hi Ben and all,

Sorry for late response - just came back from a vacation.  I found the
problem - our newest SRAdb SQLite file was not copied to the web
server due to permission issue.  I have fixed it:

> sraConvert(c('SRA036600','SRA049463','ERA062401'), sra_con= sra_con)
   submission     study    sample experiment       run
1   ERA062401 ERP000941 ERS066098  ERX024719 ERR047656
2   ERA062401 ERP000941 ERS066098  ERX024722 ERR047659
3   ERA062401 ERP000941 ERS066097  ERX024712 ERR047649
4   ERA062401 ERP000941 ERS066097  ERX024710 ERR047647
5   ERA062401 ERP000941 ERS066098  ERX024721 ERR047658
6   ERA062401 ERP000941 ERS066097  ERX024711 ERR047648
7   ERA062401 ERP000941 ERS066097  ERX024708 ERR047645
8   ERA062401 ERP000941 ERS066097  ERX024715 ERR047652
9   ERA062401 ERP000941 ERS066098  ERX024720 ERR047657
10  ERA062401 ERP000941 ERS066097  ERX024713 ERR047650
11  ERA062401 ERP000941 ERS066097  ERX024709 ERR047646
12  ERA062401 ERP000941 ERS066098  ERX024723 ERR047660
13  ERA062401 ERP000941 ERS066098  ERX024717 ERR047654
14  ERA062401 ERP000941 ERS066097  ERX024714 ERR047651
15  ERA062401 ERP000941 ERS066098  ERX024718 ERR047655
16  ERA062401 ERP000941 ERS066098  ERX024716 ERR047653
17  SRA036600 SRP006780 SRS193106  SRX062801 SRR205889

BTW, "SRA049463" is in 'unpublished' status.

Thanks for your message.  Your comments will be highly appreciated.

Jack


---------------------------------------------------------------------
Hi,

Firstly thanks to the creators of this very useful package.

I've come across SRA identifiers that don't appear to be in the database (a
minority, but still). Here's a few:

SRA036600
DRX001436
SRA049463
ERA062401
ERA062401

For example:
> library(SRAdb)
> sra_con = dbConnect(SQLite(),'SRAmetadb.sqlite')
> sraConvert(c('SRA036600'), sra_con= sra_con)
[1] submission study      sample     experiment run
<0 rows> (or 0-length row.names)

However this isn't a bogus accession because I can see it on the NCBI SRA
website.

I could be wrong but I don't think it is as simple as the metadata being
out of date because the submission dates are often relatively old
(SRA036600 was 2011-05-13) and there's metadata from more recent SRA
submissions in the SRAdb).


Any ideas?
Thanks in advance,
ben



More information about the Bioconductor mailing list