jerryji1993 / DNABERT

DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome

Home Page:https://doi.org/10.1093/bioinformatics/btab083

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Trained on mouse?

gdolsten opened this issue · comments

commented

Is there a pre-trained DNA BERT trained on mouse genome? Thanks very much,
Gabe

I guess my answer is a bit late, but like mentioned in the paper https://doi.org/10.1093/bioinformatics/btab083 the model is pre-trained on the human genome. Following the Supplementary data it is exactly "GRCh38.p13 primary assembly from GENCODE".

It is possible to pre-train your own version with a mouse genome.