Link prediction for new nodes

[faren-f]

I am using PyG for link prediction on a heterogeneous graph based on this example.
Using RandomLinkSplit(), edges can be separated into the train, validation, and test. This cross-validation method, However, only allows predicting edges whose nodes have already been seen by the training algorithm. In my case, specifically, for personalized medicine applications, I need to test my model for predicting edges for new nodes (let's consider them as new patients that were not available during the training phase). I wonder whether there is a specific way in PyG to handle such problems?