facebookresearch / esm

Evolutionary Scale Modeling (esm): Pretrained language models for proteins
MIT License
3.16k stars 627 forks source link

offset=24 in [Zero-shot variant prediction with protein language models] #680

Open yinshiyi opened 5 months ago

yinshiyi commented 5 months ago

In this run, offset is set as 24, but default as 0 It is not obvious to me why 24 is needed. Thanks! https://github.com/facebookresearch/esm/blob/2b369911bb5b4b0dda914521b9475cad1656b2ac/examples/variant-prediction/README.md?plain=1#L16C1-L16C20