Re: QSAR - Modeling biological data

Dear QSAR members,

With respect to inactive compounds, I think that it is very important to
include them in your thinking. Otherwise you run the risk of making a very
wrong prediction.

If it is easy to see how they differ from the active compounds, then a
simple "outside the scope of the QSAR" will do. If not, then you must do
you best to include them. At a minimum they should be in the prediction
set. Another approach would be to use discriminant analysis or a related
approach to investigate what properties distinguish potent from weak and
inactive compounds. These same properties might or might not be those that
appear in a final QSAR. If you have a lot of data you might try to use a
modified regression program in which you include these compounds but set
the residual of observed vs fitted to be zero if the fitted value is
greater than 128.

Yvonne Martin

jjhuusko . mappi.helsinki.fi
Sent by: qsar_society-admin(0)accelrys.com
02/08/03 05:13 AM
Please respond to qsar_society

 
        To: qsar_society{}accelrys.com
        cc:
        Subject: Re: QSAR - Modeling biological data

NB: Unless you reset the To: line, your reply goes to the entire list

---
Dear Suzanne,
Your question is justified. It seems like experimental values over
128 mg/ml shows also low activity responce? In QSARs only excact values
for activity should be used, hence I prefer to exclude the compounds
which do not have these values. Of course, some discussion of the
reasons why there are compounds which shows low activity (in this case
there might also be some analytical problems, like low solubility etc).
With all the best,
Jarmo
> Hello,
> I would like to develop a QSAR model using calculated descriptors as
> well as experimental values. The problem with the experimental values is
> the range. They are from 2 to 128 mg/ml. But some could not be
> determined and they are said to be > 128 mg/ml. How can I take into
> account in my training set of data that are greater than.
> 
> Thank you with your help
> 
> 
> Suzanne Sirois, Ph.D
> Cheminformatics, Computational Chemistry
> 
> _______________________________________________
> qsar_society mailing list
> qsar_society|,|accelrys.com
> http://ftp2.accelrys.com/mailman/listinfo/qsar_society
> 
_______________________________________________
qsar_society mailing list
qsar_society!A!accelrys.com
http://ftp2.accelrys.com/mailman/listinfo/qsar_society