ChemAxon 9682fe3ebe
11-11-2009 20:42:20
Basically, I would like to be able to index a patent by it's Markush structure and then have that structure be searchable. So my questions are:
1) How large are the enumerated libraries that are generated? In other words, what do we need in terms of disk space and CPU requirements to make this work for a large number of large Markush Structures?
2) Right now I'm pondering building a patent database in SharePoint and using Markush Search to index the database. Do you think there may be any problems with this approach (i.e. problems interfacing with SharePoint)? Also, what kind of limitations are there on using third party software to perform the search of the enumerated libraries?
3) I'm wondering what kind of limitations you have encountered with trying to represent actual patent Markush structure with Markush Search, particularly with large Makush groups.
ChemAxon a3d59b832c
17-11-2009 13:33:02
Although a specific meeting is planned later this week, and these questions will be followed up in detail, I reply here as well to have the answers for other followers of the forum.
RJohnsonUSAChemAxon wrote:
|
1) How large are the enumerated libraries that are generated? In
other words, what do we need in terms of disk space and CPU
requirements to make this work for a large number of large Markush
Structures?
|
Our Markush enumeration plugin can give the exact size of the Markush library, described by the Markush structure:
http://www.chemaxon.com/product/menum.html
However,
you do not need to worry about the disk and CPU requirements, because
we can directly search the Markush structures, without the need of
explicit enumeration.
RJohnsonUSAChemAxon wrote: |
2) Right now I'm pondering building a patent
database in SharePoint and using Markush Search to index the database.
Do you think there may be any problems with this approach (i.e.
problems interfacing with SharePoint)?
|
Currently
we are working on Sharepoint integration, which is going to be a
separate product, but is not available yet. You can also use the JChem
Base .NET API if you would like to solve integration yourself.
RJohnsonUSAChemAxon wrote: |
Also, what kind of limitations
are there on using third party software to perform the search of the
enumerated libraries?
|
We
do not use third party software, everything can be solved by ChemAxon
software. And as mentioned earlier, no enumeration of the libraries is
needed.
RJohnsonUSAChemAxon wrote: |
3) I'm wondering what kind of limitations
you have encountered with trying to represent actual patent Markush
structure with Markush Search, particularly with large Makush groups.
|
It is not the size of the
library, but rather the complexity of the Markush drawing itself is
that affects the algorithm. However, we could search very complex
Markush structures (e.g. with a library of 10^40) within reasonable timescale. (Seconds.)
I am looking forward to speak soon.
Szabolcs