DBD::mysql::db do failed

[Fang Li]

Hi all,

I am newly in PASA and I have encounter a problem when I runing in sample_data/ with the command : 

../scripts/Launch_PASA_pipeline.pl -c alignAssembly.config -C -R -g genome_sample.fasta -t all_transcripts.fasta.clean -T -u all_transcripts.fasta -f FL_accs.txt --ALIGNERS blat,gmap --CPU 2 .

When I using a database already created names as "sundaydatabase" and edited the alignAssembly.config as "MYSQLDB=sundaydatabase", it will report the error " DBD::mysql::db do failed: Can't create database 'mydatabase'; database exists ..."

Then I edited the alignAssembly.config as "MYSQLDB=newdatabase", while "newdatabase" is just a name and haven't been created indeed. The error report was "DBD::mysql::db do failed: Access denied for user 'sunday'@'%' to database 'newdatabase' ..."

I actually can login the mysql using the usrname "sunday" and with all privilege to a specific database "sundaydatabase", and using the usrname "sunday1" with the read-only privilege to this database. But I can not creat database with this two accounts.  So I want to know does PASA creat databases during its procedure and was this the reason that I am failed ?

Best,

sunday 

[Brian Haas]

Pasa does create a new database each time it is run for the initial alignment and assembly phase

-Brian

(by iPhone)

--
You received this message because you are subscribed to the Google Groups "pasapipeline-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pasapipeline-us...@googlegroups.com.
To post to this group, send email to pasapipel...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pasapipeline-users/e20a078b-d907-4b11-9b6f-cc29008946e7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Fang Li]

Thanks for that! Now I finished the two steps (alignment and comparison) with  input as a trinity fasta, a genome assembly, and a pre-existing gff file generated by homology predictor. But sorry to bother you, that I still have another question that about the gff3 output. My intention to do this is to find gene models those may be missed in the pre-existing gff, such as gene models only in this species and can not be predicted by homology prediction, but it can be  picked up through the transcript data.  While in the output gff3, I found that the genes were eighter be modified based on their previous structure or just untouched. There seemed that no newly predicted genes in it. So does PASA only deal with genes that already exist in the input pre-existing gff?  And can I still use PASA to predict the new genes that have not been found before? if yes, what script in PASA I should use?

 

Best,

Sunday

在 2016年11月30日星期三 UTC+8下午8:25:48，Brian Haas写道：

Pasa does create a new database each time it is run for the initial alignment and assembly phase

-Brian

(by iPhone)

On Nov 30, 2016, at 3:51 AM, Fang Li <lifan...@gmail.com> wrote:

Hi all,

I am newly in PASA and I have encounter a problem when I runing in sample_data/ with the command : 

../scripts/Launch_PASA_pipeline.pl -c alignAssembly.config -C -R -g genome_sample.fasta -t all_transcripts.fasta.clean -T -u all_transcripts.fasta -f FL_accs.txt --ALIGNERS blat,gmap --CPU 2 .

When I using a database already created names as "sundaydatabase" and edited the alignAssembly.config as "MYSQLDB=sundaydatabase", it will report the error " DBD::mysql::db do failed: Can't create database 'mydatabase'; database exists ..."

Then I edited the alignAssembly.config as "MYSQLDB=newdatabase", while "newdatabase" is just a name and haven't been created indeed. The error report was "DBD::mysql::db do failed: Access denied for user 'sunday'@'%' to database 'newdatabase' ..."

I actually can login the mysql using the usrname "sunday" and with all privilege to a specific database "sundaydatabase", and using the usrname "sunday1" with the read-only privilege to this database. But I can not creat database with this two accounts.  So I want to know does PASA creat databases during its procedure and was this the reason that I am failed ?

Best,

sunday 

--
You received this message because you are subscribed to the Google Groups "pasapipeline-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pasapipeline-users+unsub...@googlegroups.com.

[Brian Haas]

Hi,

If you run pasa with the parameter --TRANSDECODER, it'll attempt to annotate 'new' genes for you.  Otherwise, it relies on your existing annotation set.  Note, it'll only identify 'new' genes if the coding lengths look to be full-length (start to stop codon).

best,

~brian

On Thu, Dec 8, 2016 at 10:38 AM, Fang Li <lifan...@gmail.com> wrote:
>
> Thanks for that! Now I finished the two steps (alignment and comparison) with  input as a trinity fasta, a genome assembly, and a pre-existing gff file generated by homology predictor. But sorry to bother you, that I still have another question that about the gff3 output. My intention to do this is to find gene models those may be missed in the pre-existing gff, such as gene models only in this species and can not be predicted by homology prediction, but it can be  picked up through the transcript data.  While in the output gff3, I found that the genes were eighter be modified based on their previous structure or just untouched. There seemed that no newly predicted genes in it. So does PASA only deal with genes that already exist in the input pre-existing gff?  And can I still use PASA to predict the new genes that have not been found before? if yes, what script in PASA I should use?
>  
> Best,
> Sunday
>
> 在 2016年11月30日星期三 UTC+8下午8:25:48，Brian Haas写道：
>>
>> Pasa does create a new database each time it is run for the initial alignment and assembly phase
>>
>> -Brian
>> (by iPhone)
>>
>>
>> On Nov 30, 2016, at 3:51 AM, Fang Li <lifan...@gmail.com> wrote:
>>
>> Hi all,
>>
>> I am newly in PASA and I have encounter a problem when I runing in sample_data/ with the command :
>>
>> ../scripts/Launch_PASA_pipeline.pl -c alignAssembly.config -C -R -g genome_sample.fasta -t all_transcripts.fasta.clean -T -u all_transcripts.fasta -f FL_accs.txt --ALIGNERS blat,gmap --CPU 2 .
>>
>> When I using a database already created names as "sundaydatabase" and edited the alignAssembly.config as "MYSQLDB=sundaydatabase", it will report the error " DBD::mysql::db do failed: Can't create database 'mydatabase'; database exists ..."
>> Then I edited the alignAssembly.config as "MYSQLDB=newdatabase", while "newdatabase" is just a name and haven't been created indeed. The error report was "DBD::mysql::db do failed: Access denied for user 'sunday'@'%' to database 'newdatabase' ..."
>>
>> I actually can login the mysql using the usrname "sunday" and with all privilege to a specific database "sundaydatabase", and using the usrname "sunday1" with the read-only privilege to this database. But I can not creat database with this two accounts.  So I want to know does PASA creat databases during its procedure and was this the reason that I am failed ?
>>
>> Best,
>> sunday
>>
>> --
>> You received this message because you are subscribed to the Google Groups "pasapipeline-users" group.

>> To unsubscribe from this group and stop receiving emails from it, send an email to pasapipeline-us...@googlegroups.com.

>> To post to this group, send email to pasapipel...@googlegroups.com.
>> To view this discussion on the web visit https://groups.google.com/d/msgid/pasapipeline-users/e20a078b-d907-4b11-9b6f-cc29008946e7%40googlegroups.com.
>> For more options, visit https://groups.google.com/d/optout.
>

> --
> You received this message because you are subscribed to the Google Groups "pasapipeline-users" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to pasapipeline-us...@googlegroups.com.

> To post to this group, send email to pasapipel...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/pasapipeline-users/1b87065d-23f7-4b9b-afc2-5ff19462828b%40googlegroups.com.

>
> For more options, visit https://groups.google.com/d/optout.

--
--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas