Generalized Biological Foundation Model with Unified Nucleic Acid and Protein Language

The resources of LucaOne, including:

the model code, training scripts, embedding inference code, and trained checkpoints of LucaOne; the model code, training and evaluation scripts, datasets, trained checkpoints for downstream tasks, and additional supplementary materials.  

Due to the pre-training dataset of LucaOne is too large (1.8T), it has been deposited into CNGB Sequence Archive (CNSA) with accession number CNP0007266.

This new version updated the zip file: Supplementary.tar.gz, which contains more files and data.

Please refer to GitHub:  https://github.com/LucaOne?tab=repositories for the latest code of projects: LucaOne, LucaOneApp, and LucaOneTasks.

Since the compressed file of LucaOne CheckPoints (`TrainedCheckPoint.tar.gz`, about `33G`) is too large, it has been split into three smaller parts. Before decompressing, use the command `cat` to merge them, and then proceed with decompression.      

``` 
cat TrainedCheckPoint.tar.gz-part-* > TrainedCheckPoint.tar.gz      

tar -xzvf TrainedCheckPoint.tar.gz       
```