Graph Neural Networks using R

[LewisT1987]

Hi, hope you are well.

I want to use GNNs for a drug discovery project I will be working on, I want to use an approach similar to this paper. https://www.sciencedirect.com/science/article/pii/S0092867420301021

My issue is that I am primarily an R programmer, but most of the online examples and support for GNNs are for python. Of course, I am happy to use Python if necessary, but using R would be preferred.

Can you direct me to an example of how to use GNNs using Tensorflow/keras in R? If no examples exist, then perhaps this is an unmet functionality that needs to be addressed. Please let me know if you can help me.

Thanks Lewis

[t-kalinowski]

To my knowledge, there isn't a dedicated GNN package in R at this time, however, any functionality available in python can also easily be accessed from R via {reticulate}.

With reticulate, you can continue to work entirely in R, only needing to know enough Python to be able to read documentation (this may be helpful to quickly get up to speed: https://rstudio.github.io/reticulate/articles/python_primer.html)

[LewisT1987]

Thank you Tomasz, this is helpful.

Lewis

On Wed, Jun 21, 2023 at 2:58 PM Tomasz Kalinowski @.***> wrote:

To my knowledge, there isn't a dedicated GNN package in R at this time, however, any functionality available in python can also easily be accessed from R via {reticulate}.

With reticulate, you can continue to work entirely in R, only needing to know enough Python to be able to read documentation (this may be helpful to quickly get up to speed: https://rstudio.github.io/reticulate/articles/python_primer.html)

— Reply to this email directly, view it on GitHub https://github.com/rstudio/tensorflow.rstudio.com/issues/59#issuecomment-1601416299, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEVCPLPVMO7ACVKTXOLB5D3XMM74XANCNFSM6AAAAAAZO46YOY . You are receiving this because you authored the thread.Message ID: @.***>