annotation with Swissprot or ncbi nr ?

[Farbod Emami]

Dear All, Hi,

I have used Trinity package to assemble my De novo RNA-seq of a non model fish species. Before running GO procedure, I have used Trinotate for annotation. As it has written  in the Documentation, I have used Blastx and Blastp against swissprot database, but I have seen MANY papers that has used blastx against ncbi nr database for annotation.

my question : which is better and why so much people has routinely choose nr database blasting?

Thank you in advance

Farbod 

[Brian Haas]

swissprot contains a relatively small number of entries, but they are of high value and well annotated (arguably one of the best protein database resources available).

nr contains 'everything'.

Try looking into Trinotate and the annotation resources we make available:

http://trinotate.github.io

best,

~b

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.
To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

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

 

[Farbod Emami]

Dear Brian

Hi. I have checked the trinotate pipeline but there is no nr database blast in its inputs. I wonder if I run the blast nr how I must feed the results to Trinotate.

Thanks

On Sunday, August 30, 2015 at 6:29:02 PM UTC+4:30, Brian Haas wrote:

swissprot contains a relatively small number of entries, but they are of high value and well annotated (arguably one of the best protein database resources available).

nr contains 'everything'.

Try looking into Trinotate and the annotation resources we make available:

http://trinotate.github.io

best,

~b

On Sun, Aug 30, 2015 at 8:34 AM, Farbod Emami <farbo...@gmail.com> wrote:

Dear All, Hi,

I have used Trinity package to assemble my De novo RNA-seq of a non model fish species. Before running GO procedure, I have used Trinotate for annotation. As it has written  in the Documentation, I have used Blastx and Blastp against swissprot database, but I have seen MANY papers that has used blastx against ncbi nr database for annotation.

my question : which is better and why so much people has routinely choose nr database blasting?

Thank you in advance

Farbod 

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

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

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[Brian Haas]

It uses uniref90, which is essentially an nr databases where sequences that are less than 10% divergent are clustered and represented by a single sequence. It makes the search a lot faster due to the smaller size of the database, with minimal loss of the power of homology detection.  It's a trade-off, but a useful one.

You can learn about it here:

http://www.uniprot.org/help/uniref

If you're going to use trinotate, be sure to use the databases provided on the trinotate website, as they're tuned to the software system. 

The next version of Trinotate will be more flexible wrt databases used.

best,

~b

On Sun, Aug 30, 2015 at 10:17 AM, Farbod Emami <farbo...@gmail.com> wrote:

Dear Brian

Hi. I have checked the trinotate pipeline but there is no nr database blast in its inputs. I wonder if I run the blast nr how I must feed the results to Trinotate.

Thanks

On Sunday, August 30, 2015 at 6:29:02 PM UTC+4:30, Brian Haas wrote:

swissprot contains a relatively small number of entries, but they are of high value and well annotated (arguably one of the best protein database resources available).

nr contains 'everything'.

Try looking into Trinotate and the annotation resources we make available:

http://trinotate.github.io

best,

~b

On Sun, Aug 30, 2015 at 8:34 AM, Farbod Emami <farbo...@gmail.com> wrote:

Dear All, Hi,

I have used Trinity package to assemble my De novo RNA-seq of a non model fish species. Before running GO procedure, I have used Trinotate for annotation. As it has written  in the Documentation, I have used Blastx and Blastp against swissprot database, but I have seen MANY papers that has used blastx against ncbi nr database for annotation.

my question : which is better and why so much people has routinely choose nr database blasting?

Thank you in advance

Farbod 

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

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

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

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

 

--

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

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

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.