You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While I did put ways for having CoRE-ATAC function with other genomes, it has not been evaluated for other organisms. Differences in transcription factors and binding site sequences are the main issue here and I predict that mammals would perform better than non-mammals, especially for insulator predictions (CTCF).
I can see two strategies if using CoRE-ATAC:
Training a new model either using the organism of interest in mind, or a model using multiple organisms if training data for the specific organism is absent.
Extracting the ATAC-seq signal-based model that minimally uses DNA sequence for its predictions.
Hello developers,
Very cool tool and great paper.
I was wondering if this tool could be adapted for non-human samples (e.x. cattle)? Or was it specifically developed or trained with hg19 in mind?
Thanks so much!
The text was updated successfully, but these errors were encountered: