Query regarding practical use of delaney regression model and how well will it predict the solubility value on new drugs.
adithyaan-creator opened this issue · comments
@dataprofessor How well will the current regression model perform on new drugs? On what type of data points(new chemicals) do you think from your perspective the model will perform badly?
Great question, to answer that question we need to perform the
"applicability domain" analysis. This can be done by using a PCA scores
plot to see whether the new compound falls within the boundaries of the
training set compounds or not.
…On Thu, Oct 8, 2020 at 1:42 AM Adithya ***@***.***> wrote:
@dataprofessor <https://github.com/dataprofessor> How well will the
current regression model perform on new drugs? On what type of data
points(new chemicals) do you think from your perspective the model will
perform badly?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#3>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AMLTBY3WLHQQBU4OZTN627TSJSZBBANCNFSM4SHYO5EQ>
.
What are some other places where I can use Machine Learning algorithms in the drug discovery pipeline? Saw one on using RNN for generating SMILE notations.
That's a great question, actually there are so many use cases, and yes amongst that is to generate SMILES notation. One can also apply ML to explore the entire proteome and perform network analysis to visualize the complex protein-protein interactions. Another is to perform drug repurposing of existing drugs for treating a new disease.