Workflow examples that illustrate the ChEMBL RESTful web services

KNIME Analytics Platform
1

Using the ChEMBL web services is a good way to fetch and integrate ChEMBL data without having to install SQL dumps locally or write SQL queries.

I am attaching here a set of workflows that conveniently wrap the functionality of these web services in KNIME.

The workflows can serve as building blocks for more sophisticated and information-rich data analysis. See for example the NN_Activity_Profile workflow attached.

I mainly used the XML reader and the XPath nodes to fetch the URL and tabulate the XML response.

Any comments are welcome.

 

Regards,

George Papadatos

ChEMBL group

2

See also: http://chembl.blogspot.co.uk/2012/07/access-to-chembl-web-services-via.html

3

Is it worth having these uploaded to KNIMES Public Workflow Server.


Simon.

4

Hi Simon,

Sure. How does one do this?

George

5

Thank you for such nice examples. I have taken the liberty of uploading them to the public example server under the community section (/99_Community/08_ChEMBL_WebServices/).  

Regards,

Aaron

6

Great. Many thanks Aaron.

George

7

Thanks for sharing and uploading these workflows. I think these are very useful. How would one get the structures for e.g. target + bioactivities.

Best,

Natasja

8

Hi Natasja,

Here you go:

https://dl.dropbox.com/u/22273283/Compound%2BAssay%2BTarget_Bioactivities_NEW.zip

Please note that if you have more than 1000 bioactivities for a target or assay, it is not a good idea to fetch the compound information. In such case, try first to filter out bioactivities (e.g. keep only IC50 values) and then retrieve the compound information.

 

George

ChEMBL group

9

In KNIME 2.6 a stricter URL time out limit was introduced.

Please add the following line to your knime.ini file to avoid your ChEMBL web service requests timing out too early.

-Dknime.url.timeout=6000

 

Regards,

George

10

Dear George,

 

thank you very much for the updated workflow - I played around with, but I cannot figure out how to dig out the structural data, after I filter out quite a large amount of data..

Cheers & thanks in advance,

Paul

11

Hallo Paul,

In your example, just edit the last joiner node by changing the joining columns:

left table --> ingredient_cmpd_chemblid

right table --> chembl_id

The rest should stay the same and it should work!

Best regards,

George

 

12

Hello again,

I changed your workflow again to make it more general and more efficient.

Please use this version and let me know if it works as expected.

 

George

13

Dear George,

thanks a lot for the updated workflow, it works perfect!

Cheers & Thanks,

Paul

14

Hi George,

I was wondering whether the example workflows are still available as I'm having problems accessing them in dropbox.

Thanks!

Barry

15

Hi,

I just downloaded the zip file above and imported it to KNIME. Fresh installation of KNIME with all updates.

When I run the first node "FETCH" I get the following error:

WARN      GroupBy                            No aggregation column defined
WARN      BrowserInitializer                 Found WebKit version 2.2.3. Only WebKit version 1.x is supported.

ERROR     XML Reader                         Execute failed: Read timed out

Are there other things that must be installed, before I can run it?

 

I tried to fix it by add

-D.knime.url.timeout=100000

to knime.ini

but it did not solve the problem.

best