search

WGLab / DeepMod

DeepMod: a deep-learning tool for genomic-scale, strand-sensitive and single-nucleotide based detection of DNA modifications
Other
97 stars 35 forks source link

No event in fast5 files. #59

Open Bin-Ma opened 1 year ago

Bin-Ma commented 1 year ago

Hi @liuqianhn, I want to detect the 6mA methylation from Guppy basecalled FAST5 files. Our FAST5 files have been splited using multi_to_single_fast5. However, there are no event in all FAST5 files. Is there any way to append the event to the fast5 file?

/Group /Analyses Group /Analyses/Basecall_1D_000 Group /Analyses/Basecall_1D_000/BaseCalled_template Group /Analyses/RawGenomeCorrected_000 Group /Analyses/RawGenomeCorrected_000/BaseCalled_template Group /Analyses/RawGenomeCorrected_001 Group /Analyses/RawGenomeCorrected_001/BaseCalled_template Group /Raw Group /Raw/Reads Group /Raw/Reads/Read_17756 Group /Raw/Reads/Read_17756/Signal Dataset {19660/Inf} /UniqueGlobalKey Group /UniqueGlobalKey/channel_id Group /UniqueGlobalKey/context_tags Group /UniqueGlobalKey/tracking_id Group

I have tried the albacore in base calling, but the flowcell number and kit number is not available from the company. Looking forware to your reply.

Bin

liuqianhn commented 1 year ago

@Bin-Ma DeepMod does not support guppy. Please check https://github.com/WGLab/DeepMod2 for the latest version.

Bin-Ma commented 1 year ago

@liuqianhn Thanks for your reply. I retry the step of base calling and it works now. However, the correlation against other software (Deepsignal, mCaller) was low. I don't know how to balance it.

图片

Looking forward to your reply. Best, Bin

liuqianhn commented 1 year ago

@Bin-Ma DeepMod was NOT updated since 2019, and also NOT tested against the latest Albacore v2.3.* and Guppy. This could be an issue.

kaichop commented 1 year ago

Can you try to do a Albacore v1 basecalling, and then assess the results? The original model was trained on Albacore v1 so it will not work optimally on v2, which was described in the DeepMod GitHub page.

Bin-Ma commented 1 year ago

@liuqianhn @kaichop Thanks for your reply! I will try to install the Albacore v1 for base-calling recently. Here, I have sequenced the E. coli genome using nanopore with about 20X~200X and try to detect the methylation of E. coli. However, I have no idea about how to set the cut-off value of each software. How many 6mA/5mC methylations are available on the E. coli genome? Whether the sequence depth in this experiment is accecptable? Looking forward to your reply. Best, Bin

kaichop commented 1 year ago

The E. coli DNA from Simpson was either untreated or treated with the CpG methyltransferase M.SssI. In terms of expectation on a new genome, these papers may help https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4231299/ and https://link.springer.com/chapter/10.1007/978-1-349-06343-7_32

On Tue, Mar 21, 2023 at 5:07 AM Bin-Ma @.***> wrote:

@liuqianhn https://github.com/liuqianhn @kaichop https://github.com/kaichop Thanks for your reply! I will try to install the Albacore v1 for base-calling recently. Here, I have sequenced the E. coli genome using nanopore and try to detect the methylation of E. coli. However, I have no idea about how to set the cut-off value of each software. How many 6mA/5mC methylations are available on the E. coli genome? Looking forward to your reply. Best, Bin

— Reply to this email directly, view it on GitHub https://github.com/WGLab/DeepMod/issues/59#issuecomment-1477486907, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNG3OA4O54TB46KCNA3WRLW5FVWVANCNFSM6AAAAAAU3CFAYU . You are receiving this because you were mentioned.Message ID: @.***>

Bin-Ma commented 1 year ago

@kaichop Thanks for your help!

Best, Bin