MSA for RNA is this available?

[Hannns22]

Hello,

First of all, thank you for providing such a great source code for my research. I have a question regarding the inference process. As I noticed in the code, protein data can be used as input to the model by linking the msa_directory, allowing the model to work with MSA. However, I understand that for RNA, this functionality to use MSA as an input is not yet supported. Is that correct?

If so, do you have any updates on when this feature might be added? or can I train RNA MSA from scratch using train_demo.sh?

Thank you!

[zhangyuxuann]

@Hannns22 Yes, MSA for RNA is not available yet . Although Protenix have already support most of the code for RNA MSAs，due to the hit rate of RNA is very low, we did not find a significant improvement in our experiments. In order to simplify the whole pipeline, we did not use RNA MSA for both training and inference in our released model.

if you want to provide RNA MSA for training your own model, you can ref to

Protenix/protenix/data/msa_featurizer.py

Line 502 in 80d8c5f

class RNAMSAFeaturizer(BaseMSAFeaturizer):

Protenix/configs/configs_data.py

Line 191 in 80d8c5f

"rna": {