NCIT-PDX Finder Ontolia

[NatConte]

In PDX Finder we use NCIT to map drugs and regimen. We needed a way to map the correct regimens to their corresponding individual drugs to power our search. Csaba Halmagyi in my team developed ONTOLIA to overcome this issue and we thought you may be interested by reusing the tool.

I have attached a ppt describing the tool fyi

cc'in @dosumis

Happy to discuss further
Nathalie Conte
PDXFinder lead

[decorons]

Thanks, we will take a look and possibly get back to you!! Sherri de Coronado, NCI CBIIT

[dosumis]

Hi @NatConte,

I think it would be great if NCIT could encode the regimen to drug mappings you've made as ontology relationships. As I understand it, the ONTOLIA tools has broader ambitions, so maybe a simple mapping table would be a good place to start.

(BTW - I don't see the attachment - perhaps best to link to slides hosted somewhere?)

Cheers,
David

[NatConte]

Thanks @decorons and @dosumis
here is the presentation link - https://drive.google.com/file/d/1-hwzMj-3UwV7R6VmFQT-DiUgRJte0L32/view?usp=sharing
Csaba Halmagyi in my team developped the tool, he is currently on holidays , he can produce a table when back.
cheers

[mellybelly]

@NatConte I think that the GA4GH clinPheno workstream may be initiating a standard for treatment course that you might be interested in participating in. We also want to be able to work on coordinating course modeling with the therapeutic response relationship to variants: ga4gh/va-spec#31

info on the workstream is here: https://ga4gh-cp.github.io/

@monicacecilia can you help us coordinate? Also relates to the new CCDH center. @balhoff

[NatConte]

thanks for the information - this is a very interesting initiative, happy to contribute ! In PDX Finder, we capture high level treatment/response from patient treatment and drug dosing studies (treatment on PDX mouse models) - We are also working into standardising PDX drug dosing studies from preclinical models of cancer - patient and model standardisation efforts have a synergistic potential to better integrate data.

about the variant relative resistance to drug this is extremely valuable and some databases have some info about variant specific resistance to drug this, Civic db, COSMIC, VICC - I am sure you are pretty familiar with those.

Happy to discuss further.

[NatConte]

Hi @dosumis @mellybelly @decorons ,

Attached a mapping table - showing mappings using ontolia on 3750+ drug regimens in NCIT
columns shows NCIT term url (A) | NCIT regimen label (B) | NCIT regimen synonyms (C)| and Ontolia mapped individual drugs labels (D)

https://drive.google.com/file/d/1euI48aPOjDFdm_ZDRQnns3iHOdjjAn0d/view?usp=sharing

Feel free to review and comment.
cheers

[CsabaHalmagyi]

Hi everyone,

Nathalie already introduced me. Should you have any technical question regarding the algorithm, I am happy to answer.
Regards,
Csaba

[NatConte]

hi all, @decorons @dosumis @CsabaHalmagyi @mellybelly
Did you have a chance to review our mapping files? is this something that could be used on the NCIT side to update synonyms or create new axioms? let us know if you have any comments
Thanks
Nathalie