Have you evaluated the transferability of NRI-MD?

[lanjinraol]

How about the performance of NRI-MD when training dataset and test dataset are the trajectories of different protein?

[juexinwang]

That is a good question. To the best of my knowledge, the transferability of the model is unknown. Usually, it is low, there are limited theoretical experiences to support the model can be training on one protein and test on another one