RE: oxygen a "chemical drug"?

10 views
Skip to first unread message

Yannick Pouliot

unread,
Jun 14, 2007, 12:03:45 PM6/14/07
to Jyothi Paniyadi, stanf...@googlegroups.com

> -----Original Message-----
> From: Jyothi Paniyadi [mailto:sup...@ingenuity.com]
> Sent: Wednesday, June 13, 2007 8:39 PM
> To: ypou...@stanford.edu
> Subject: Re: oxygen a "chemical drug"?
>
> Dear Dr. Pouliot,
>
> Thank you for contacting Ingenuity Systems customer support. "Oxygen"
> in IPA is considered an endogenous chemical. Variants of oxygen,
> however, may be classified under different categories. For eg. oxygen
> radical comes under the category of chemical toxicant. You can get this
> information by typing oxygen in the search field and looking up the
> "Type" field.
>
> Regarding your second question, using the Path Explorer feature, user
> can obtain a) shortest path b) shortest path+1 node c) shortest path+2
> nodes. For the shortest path, IPA first tries to look up an interaction
> involving just the selected nodes (direct or indirect interactions). If
> such a connection is not available, it then looks for connections
> involving intermediate node/s as specified by the user (by selecting
> "molecule types"). In your case, a shortest path without any
> intermediate node does exist between TP53 and DHFR. This is what is
> being shown in the following relationship.
>
> "In cultured transformed fibroblasts expressing RAS [HRAS] protein and
> MYC protein from embryonic mouse, low oxygen stress of cultured
> transformed fibroblasts expressing RAS [HRAS] protein and MYC protein
> from embryonic mouse increases repression of mouse Dhfr gene that
> involves mouse P53 [Trp53] protein. "
>
> This is an indirect interaction between TP53 and DHFR and NOT an
> interaction involving an intermediate node.
>
> For paths involving the intermediate node "chemical Drug", see
> shortest path+1 node or shortest path+2 nodes.

Thanks Jyothi. This is understood. Looking at the image I sent (appended),
the confusion is this: I'm asking for any relationships between TP53 and
DHFR involving a chemical drug, and I get one path back. Great! Problem is,
I don't see a drug being involved in this relationship, thus my question
about whether oxygen is considered a drug here. Q: *Is* there a drug
involved in this relationship?

Cheers,

Yannick

>
> Please let me know if this answered your question or if you require
> further clarification.
>
> Best regards,
>
> Jyothi Paniyadi,Ph.D
> Customer Support
> Ingenuity Systems
> 1700 Seaport Blvd.
> 3rd Floor
> Redwood City, CA 94063
> 650.381.5111 phone
> 650.963.3399 fax
> Sup...@ingenuity.com
> www.ingenuity.com
>
>
>
> _______________________________________________
> Sent from Salesforce. experience success.
> http://www.salesforce.com/

ChemDrug1.jpg
Message has been deleted

Chris Kirchberg

unread,
Jun 15, 2007, 12:49:54 PM6/15/07
to StanfordIPA
Hi Yannick,

To answer your question: No, there is no chemical drug involved in
this relationship. Since IPA found a connection from TP53 to DHFR
without adding any nodes (chemical drug nodes in this case), it will
not force adding a chemical drug as a node. If one were to goto the
shortest path +1 or +2, there are no chemical drug nodes that IPA
could add to connect the two.

However, if you choose the direction between the two to be any
direction, then there will be 4 chemical drug nodes added for shortest
path +1 and two additional nodes if you choose shortest path +2.

Cheers,

Chris Kirchberg
Ingenuity Systems

> > Supp...@ingenuity.com


> >www.ingenuity.com
>
> > _______________________________________________
> > Sent from Salesforce. experience success.
> >http://www.salesforce.com/
>
>
>

> ChemDrug1.jpg
> 956KViewDownload

Yannick Pouliot

unread,
Jun 15, 2007, 12:53:29 PM6/15/07
to stanf...@googlegroups.com
Thanks for a very clear answer, Chris. Response below.

> -----Original Message-----
> From: stanf...@googlegroups.com
> [mailto:stanf...@googlegroups.com] On Behalf Of Chris Kirchberg
> Sent: Friday, June 15, 2007 9:45 AM
> To: StanfordIPA
> Subject: Re: oxygen a "chemical drug"?
>
>

> Hi Yannick,
>
> To answer your question: No, there is no chemical drug involved in
> this relationship. Since IPA found a connection from TP53 to DHFR
> without adding any nodes (chemical drug nodes in this case), it will
> not force adding a chemical drug as a node.

Huh! My interpretation of this behavior is that this violates what the user
requested: find any paths that involve a chemical drug. The system returned
the only path it could find w/o taking into account the constraint for a
chemical drug. Or am I missing something as to why it is behaving this way?

> If one were to goto the
> shortest path +1 or +2, there are no chemical drug nodes that IPA
> could add to connect the two.

OK. Wouldn't the intuitively obvious answer be "0 paths"?

Much appreciated!

Cheers,

Y

>
> However, if you choose the direction between the two to be any
> direction, then there will be 4 chemical drug nodes added for shortest

> path +1 and two additional nodes if you choose shortest path +2. See
> screen shot attached.


>
> Chris Kirchberg
> Ingenuity Systems
>
> On Jun 14, 10:03 am, "Yannick Pouliot" <ypoul...@stanford.edu> wrote:

> > > Supp...@ingenuity.com


> > >www.ingenuity.com
> >
> > > _______________________________________________
> > > Sent from Salesforce. experience success.
> > >http://www.salesforce.com/
> >
> >
> >

> > ChemDrug1.jpg
> > 956KViewDownload
>
>
>

Chris Kirchberg

unread,
Jun 15, 2007, 1:58:05 PM6/15/07
to StanfordIPA
Hi Yannick,

You are not missing anything. I will put in a request to restrict
connections only to those node types chosen and not to connect if the
nodes do not exist in the current pathway or could not add.

Thanks for your input.

Chris

Yannick Pouliot

unread,
Jun 15, 2007, 7:18:29 PM6/15/07
to cjkir...@gmail.com, stanf...@googlegroups.com

> -----Original Message-----
> From: stanf...@googlegroups.com
> [mailto:stanf...@googlegroups.com] On Behalf Of Chris Kirchberg
> Sent: Friday, June 15, 2007 10:58 AM
> To: StanfordIPA
> Subject: Re: oxygen a "chemical drug"?
>
>
> Hi Yannick,
>

> You are not missing anything. I will put in a request to restrict
> connections only to those node types chosen and not to connect if the
> nodes do not exist in the current pathway or could not add.
>
> Thanks for your input.

You're welcome. Might it be possible for the feature request number to be
communicated back to the list to facilitate referencing going forward? Yes,
this is awfully techy :-)

Cheers and have a good weekend,

Y

Reply all
Reply to author
Forward
0 new messages