akundaje
commented
2 years ago Hi Changxu
The BPNet models for ATAC/DNase are not yet finalized. We had to make significant changes to the bias correction component which took a lot longer than expected. But we are close. We expect a release of the code and a preprint by July or August this year (2022).
Thanks Anshul.
On Thu, May 12, 2022 at 11:00 AM ChangxuFan @.***> wrote:
Dear developers,
This message comes from a very confused graduate student. Briefly, I'm trying to adopt the bpnet framework to my ATAC and Dnase-seq data (to avoid eyeballing motif scan results). However, I'm not sure what's the recommended practice to do so. From Anshul's early response in this issue (#18 (comment) https://github.com/kundajelab/bpnet/issues/18#issuecomment-771203232), it seems that the basepairmodels repo should contain such info. However, the tutorial offered in this repo was based on ChIP-seq. I found another repo called kerasAC, which seems to be developed for ATAC/Dnase-seq data, as implied by the "AC" in its name. However, I had some issues getting the repo to work, and the examples offered are not as detailed as this repo. In summary, I'm not entirely sure which is the "right" repo to use for ATAC/Dnase data based analyses and if there is a guideline for parameters. I would highly appreciate it if you could point me to the correct direction!
Thank you so much, Changxu Fan
— Reply to this email directly, view it on GitHub https://github.com/kundajelab/bpnet/issues/41, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABDWEI7GBESLFQBPIOIGZTVJVBNLANCNFSM5VZBIBAQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>
zhenzuo2
DongzeHE
Dear developers,
This message comes from a very confused graduate student. Briefly, I'm trying to adopt the bpnet framework to my ATAC and Dnase-seq data (to avoid eyeballing motif scan results). However, I'm not sure what's the recommended practice to do so. From Anshul's early response in this issue (https://github.com/kundajelab/bpnet/issues/18#issuecomment-771203232), it seems that the
basepairmodelsrepo should contain such info. However, the tutorial offered in this repo was based on ChIP-seq. I found another repo calledkerasAC, which seems to be developed for ATAC/Dnase-seq data, as implied by the "AC" in its name. However, I had some issues getting the repo to work, and the examples offered are not as detailed as this repo. In summary, I'm not entirely sure which is the "right" repo to use for ATAC/Dnase data based analyses and if there is a guideline for parameters. I would highly appreciate it if you could point me to the correct direction!Thank you so much, Changxu Fan