Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[bioontology-support] [BioPortal] Feedback from agravina

John Graybeal jgraybeal at stanford.edu
Mon Mar 30 08:46:20 PDT 2020


Hello Alessio,

I am sorry your work is taking so long. I can give you some suggestions you can try right away, but because you are dealing with a very large corpus, the virtual appliance definitely will be your best option.

The key way to accelerate the process is by choosing only the most important ontologies to use for annotation.  (This means the Annotator won't have to look up annotations in the other 900+ ontologies in the system.)  Also, by setting the options Match longest only, Exclude numbers, and Exclude synonyms I believe you will minimize annotation time; but I have not confirmed this.

Since system load on the BioPortal side will strongly affect annotation rate, you can monitor the response time for your calls and make sure you aren't overloading the system to the point of inefficiency. But keeping in mind there are millions of API calls per day to BioPortal, I think trying to use BioPortal as a high-speed annotation system will not be a satisfying user experience for yourself or its other users.

The OntoPortal Virtual Appliance is described at https://www.bioontology.org/wiki/Category:NCBO_Virtual_Appliance. It is available in both a form deployable on your own systems (version 2.5), and in an AWS AMI (version 2.4). To apply for access, follow the instructions in that document. You will have to download into your Virtual Appliance the ontologies that you want to use for annotation.

In a few weeks (by the end of April), we expect to release a version 3.0 of the OntoPortal Virtual Appliance, which is somewhat easier to set up, includes additional annotation capabilities with the AnnotatorPlus, and is available in both local and AWS formats.

Because these systems run only on your own dedicated hardware, they will likely be much faster than BioPortal. (And of course, you can increase that speed by getting faster or more hardware, so you have much more control over the system.)

I hope this information helps you with your work.

John

On Mar 30, 2020, at 2:26 AM, support at bioontology.org<mailto:support at bioontology.org> wrote:


Name: agravina

Email: gravina.alessio at gmail.com<mailto:gravina.alessio at gmail.com>

Location: http%3A%2F%2Fbioportal.bioontology.org<http://2Fbioportal.bioontology.org>%2Fresource_index

Feedback:

reply to: gravina.alessio at gmail.com<mailto:gravina.alessio at gmail.com>

Hi,
I am annotating a dataset with your NCBO annotator through REST API. Unfortunatelly, with REST API I am able to annotate 8 sentences per second, on average. By making some statistics this process will require 4 years.
So I am asking to you a suggestion to speed up the process. I have read of an AWS virtual appliance, do you think it could be beneficial?
Is the AWS virtual appliance a local version of the web service? or does the AWS virtual appliance query the web service though REST API?
Do you have other suggestion to speed up the process?
Thank you for your help.
Kind regards,
Alessio

_______________________________________________
bioontology-support mailing list
bioontology-support at lists.stanford.edu<mailto:bioontology-support at lists.stanford.edu>
https://mailman.stanford.edu/mailman/listinfo/bioontology-support

========================
John Graybeal
Technical Program Manager
Center for Expanded Data Annotation and Retrieval /+/ NCBO BioPortal
Stanford Center for Biomedical Informatics Research
650-736-1632  | ORCID  0000-0001-6875-5360



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/bioontology-support/attachments/20200330/ffd3085e/attachment-0001.html>


More information about the bioontology-support mailing list