Hi Hiram,

Sorry to get back to you so late. Haven't had time to test if it is trackDb
table problem. Now, I have another question about the genomic position on
the genome browser.

I spot that some of the codons are not consistent with the DNA sequence.

For example, at chr10:91,404,661-91,404,663, the DNA sequence shows CGG and
should be "R", but the codons on both UCSC and RefSeq are "P".

Could help me with the problem first?

Thank you very much!


Sincerely,
Woody

2010/11/2 Hiram Clawson <[email protected]>

> Good Morning Woody:
>
> Do you know where you got the trackDb table that only has 134 rows in it ?
> How do you create the trackDb table ?  If you fetch our hg19 trackDb table,
> you should have at least 300 rows.
>
> We have a lot of mysteries here and I am thoroughly puzzled.  We are going
> to need to get some of the kent source tree command built.  Is your build
> functioning in your source tree ?  Can you go to the directory
> src/hg/hgsql and build the hgsql command.  Set up your $HOME/.hg.conf
> file with the instructions from src/product/README.mysql.setup
> and get the hgsql command functioning so you don't need to type
> user names and passwords with your mysql command.  We also need
> $HOME/.hg.conf
> functioning so we can use other kent commands.  Get the command hgTrackDb
> built in the directory src/hg/makeDb/hgTrackDb/ and hgFindSpec
> in src/hg/makeDb/hgFindSpec
> There may be other commands we need but this is a good start.
>
> Following the instructions in src/product/README.building.source
> to get the source tree built.
>
> I had assumed you have been using the source tree all along.  Do you
> build the CGI binaries or copy ours ?
>
> --Hiram
>
> ----- Original Message -----
> From: "Woody Lin" <[email protected]>
> To: "Hiram Clawson" <[email protected]>
> Cc: [email protected]
> Sent: Monday, November 1, 2010 10:51:31 PM GMT -08:00 US/Canada Pacific
> Subject: Re: [Genome] Question about installing genome browser (error:
> wgEncodeGisDnaPet)
>
> Dear Hiram,
>
>
> It looks like you are using hgsql instead of mysql, so I changed hgsql to
> mysql and got the following messages:
>
>
>
> My-Server: trackDb admin$ mysql -u root -p -e "select count(*) from
> trackDb;" hg19
> Enter password:
> +----------+
> | count(*) |
> +----------+
> | 134 |
> +----------+
> My-Server: trackDb admin$ make GIT=echo EXTRA=-strict DBS=hg19 alpha
> find . -type l -exec rm {} \;
> echo pull
> pull
> ./loadTracks -release=alpha trackDb hgFindSpec hg19
> ./loadTracks: line 58: hgsql: command not found
> Note: database hg19 does not exist, skipping
> ./checkMetaDb alpha metaDb hg19
> ./checkMetaDb: line 22: hgsql: command not found
> Note: database hg19 does not exist, skipping
>
>
> Woody
>
> 2010/11/2 Hiram Clawson < [email protected] >
>
>
> Good Morning Woody:
>
> Please try this command in your source tree directory
> src/hg/makeDb/trackDb/
>
> $ hgsql -e "select count(*) from trackDb;" hg19
> $ make GIT=echo EXTRA=-strict DBS=hg19 alpha
> $ hgsql -e "select count(*) from trackDb;" hg19
>
> This load of the trackDb table with the -strict argument
> will load only tracks that actually have tables in your hg19 database.
> You should note the count of entries in the trackDb table will be
> lower after the 'make' command.
>
> --Hiram
>
>
>
>
> Woody Lin wrote:
>
>
> By the way, currently, I have only imported two databases "hg19" and
> "hgFixed". (all tables are imported)
> Do I need extra databases?
>
>
_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to