Jin Su

Hi, The higher the mutational effect score is, the better our model thinks such mutation should be. You can sort these rows by the mutational effect on a descending order....

I think theoretically the range of the mutational effect score is from -∞ to +∞. You could check the definition of mutational score [here](https://github.com/westlake-repl/SaprotHub/wiki/SaprotHub-v2-(latest)#introduction-of-saprot-mutation-score). In my opinion, I think +6...

Hi, You could follow this tutorial https://github.com/westlake-repl/SaProt?tab=readme-ov-file#fine-tune-saprot to finetune SaProt. You could run the GO dataset by replacing the ``.yaml`` file with the GO task :)

Hi! Sorry we also don't know how to convert the label ids to original text annotations. The GO and EC datasets we downloaded had already been processed into digital labels.

Hi, I think it's due to the imcompatibility with the version of pytorch-lightning. Could you degrade your pytorch-lightning to 1.8.3?

If you don't want to record your training then set ``logger`` to ``False`` should work. The error ``No supported gpu backend found!`` seems to be caused by your hardware configuration....

The problem is more likely due to the ``sbatch`` command, not wandb. I'm not familiar with slurm. Perhaps you could check whether ``sbatch`` does some additional operations that conflicts with...

Hi, The beta-lactamase dataset only contains mutation variants of a wild type protein. Structures for these variant are not available. We trained SaProt only based on pure sequence.

Yes :)

Hi, thank you for your interest in our work and asking some intriguing questions! > When I run this code for MLM finetuning, ``python scripts/training.py -c config/Pretrain/saprot.yaml``. I only need...