Detect other modification and train model-5-Bromo-dCTP and 5-Iodo-dCTP

[santoshbiowarrior333]

Dear remora team,

Thank you for this amazing tool, i am new to this.
I am looking for a way to detect the 5-Bromo-dCTP and 5-Iodo-dCTP.
I have added these in order at specific strand stretch.
IN terms of samples, i have no incoproration sample, and another with Bromo around a known region, then another sample with iodo around known region. THen third sample as Brd then iodo in sequence.

Is it possible to train the model based on these and analyze genomic data?

Thank you.

Best,
Dr. Santosh

[marcus1487]

Could you explain in a bit more detail exactly the nature of the incorporation? Specifically, do you know specific locations where the modified base will be deposited? Within this region is the modified base completely substituted or only substituted in some reads at some of the positions within the read? The provided Remora interface is only intended for samples where all reads covering a reference position are known to be modified (e.g. bacterial DNA methylation).

The other main point to consider here is the sequence and modified base content for both the training and inference. Remora models consume a small region of signal ~100-600 signal points (or about 12 signal points per base from current release DNA settings). If these windows of signal and sequence training data are not sufficiently representative of the desired inference data then the model is likely to perform poorly. Having a better understanding of the technical details of the proposed training and inference data will help a lot in advising on the feasibility of the proposed experiment.

[santoshbiowarrior333]

Hi Marcus, Thank you for your response. To give a quick context, these modifications will be incorporated in the plasmid DNA by polymerase invitro. I know the starting point where it will be incorporated and will be in a region. I will have samples in which these combinations will be available (1) alone brdC, (2)idC, (3) both at 2 different regions, (4) no incorporation. Plasmid is obviously methylated. No all the reads may not be modified in the process like methylation. But in the reads which has modifications (all C in that range will be modified). Length of the DNA would be around 5kb. Range for modifications region would be 200bp. Hope it gives a bit of insight. And I am happy to provide more info. Thank you very much. Best, Santosh.

…

________________________________ From: Marcus Stoiber ***@***.***> Sent: Thursday, July 24, 2025 9:51 PM To: nanoporetech/remora ***@***.***> Cc: Dr. Santosh_kumar ***@***.***>; Author ***@***.***> Subject: Re: [nanoporetech/remora] Detect other modification and train model-5-Bromo-dCTP and 5-Iodo-dCTP (Issue #219) [https://avatars.githubusercontent.com/u/13755444?s=20&v=4]marcus1487 left a comment (nanoporetech/remora#219)<#219 (comment)> Could you explain in a bit more detail exactly the nature of the incorporation? Specifically, do you know specific locations where the modified base will be deposited? Within this region is the modified base completely substituted or only substituted in some reads at some of the positions within the read? The provided Remora interface is only intended for samples where all reads covering a reference position are known to be modified (e.g. bacterial DNA methylation). The other main point to consider here is the sequence and modified base content for both the training and inference. Remora models consume a small region of signal ~100-600 signal points (or about 12 signal points per base from current release DNA settings). If these windows of signal and sequence training data are not sufficiently representative of the desired inference data then the model is likely to perform poorly. Having a better understanding of the technical details of the proposed training and inference data will help a lot in advising on the feasibility of the proposed experiment. — Reply to this email directly, view it on GitHub<#219 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AS6RYW76JMVQEAHPM7RXCFL3KFBMJAVCNFSM6AAAAACA6Z7NHWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCMJUHA4TSOJWHE>. You are receiving this because you authored the thread.Message ID: ***@***.***>