Any suggestions what are the good Python models/packages to predict small molecule (SMILES input format) binding to proteins ?

Any further information - papers, yours experience, whatever - is welcome !

PS

I am aware of: Chembert - (created some notebook - example for beginners): https://www.kaggle.com/code/alexandervc/chembert2a-smiles-embeddings-for-beginners

Siamese-smole-bert - (created some notebook - example for beginners): https://www.kaggle.com/code/alexandervc/siamese-smole-bert-smiles-embeddings-for-beginners

There is also Chemprop https://chemprop.readthedocs.io/en/latest/tutorial.html But have not tried yet.

Some survey of different SMILEs embeds: https://medium.com/p/93246c75be6a

PSPS

The context of the question - current Kaggle challenge: https://www.kaggle.com/competitions/leash-BELKA/code To predict small molecule binding to three proteins Dataset includes 90 millions small molecules - mainly non-binders.