No different colors in 5C methylation representation

[vyx-lucy-kaplun]

Hi,
There is a problem with representation of methylation: 2-color system for 5mC (as well as any other system) displays all called 5C the same color. It is either all blue, all red, or all pink, but not different colors for methylated and not-methylated residues, as it is done in a desktop IGV version. Therefore, there is no way to see what is methylated and what is not. Is there a way to fix it? Thank you.

[jrobinso]

I don't understand what you are asking, perhaps a screenshot would help. The colors and rules for applying them for base modifications are the same on IGV desktop and igv.js. Also, "5C" is not a call, did you mean 5mC?

[vyx-lucy-kaplun]

My apologies, I should have attached screenshots right away. Here are two images to illustrate the difference: same bam file, same area, same settings (2-color, all).

[jrobinso]

What version of igv.js are you using? I cannot reproduce that behavior, this is what I see with my test dataset, which contains 5mC and 5hmC calls.

[vyx-lucy-kaplun]

We are on v. 3.0.5

[jrobinso]

I will need to see your data, in addition to your track configuration. I am not able to reproduce your result. See my screenshot above. I don't need the entire BAM file, just the portion shown in your screenshot. You can use samtools to extract a slice. If needed send email to igv-team@broadinstitute.org and I will provide you with a link to a dropbox folder.

[vyx-lucy-kaplun]

Sure, thank you! E-mailing you now.

[jrobinso]

Thanks for the test data. Below is what I see with the track configuration shown. I'm not really sure what could explain the difference, but could you try the exact configuration I am using with the bam file you sent us, and also double-check your igv.js version?

  {
                    type: 'alignment',
                    format:'bam',
                    visibilityWindow: 250_000,
                    samplingDepth: 50,
                    name: 'EXAMPLE',
                    url: "IGV_5mC_issue_chr15_23646787_23648567.bam",
                    indexURL: "IGV_5mC_issue_chr15_23646787_23648567.bam.csi",
                    displayMode: 'EXPANDED',
                    height: 400,
                    maxHeight: 500,
                    alignmentRowHeight: 15,
                    colorBy: 'basemod2',
                    showDeletionText: true
                }

[vyx-lucy-kaplun]

Thank you! We will try that and let you know. Lucy

…

On Wed, Sep 18, 2024 at 10:45 PM Jim Robinson ***@***.***> wrote: Thanks for the test data. Below is what I see with the track configuration shown. I'm not really sure what could explain the difference, but could you try the exact configuration I am using (with appropriate changes to URL), and also double-check your igv.js version? { type: 'alignment', format:'bam', visibilityWindow: 250_000, samplingDepth: 50, name: 'EXAMPLE', url: "IGV_5mC_issue_chr15_23646787_23648567.bam", indexURL: "IGV_5mC_issue_chr15_23646787_23648567.bam.csi", displayMode: 'EXPANDED', height: 400, maxHeight: 500, alignmentRowHeight: 15, colorBy: 'basemod2', showDeletionText: true } Screenshot.2024-09-18.at.10.36.32.PM.png (view on web) <https://github.com/user-attachments/assets/7935d925-1631-4b58-9dbf-0f2e17877830> — Reply to this email directly, view it on GitHub <#1895 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANGYOKL5P52ZWSR6VPMONUDZXI3EJAVCNFSM6AAAAABOLXHMN2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNJZHA3DMOJUGQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- <https://www.facebook.com/Variantyx/> <https://www.linkedin.com/in/ludmilakaplun/> <https://www.instagram.com/variantyxinc/> Lucy Kaplun, Ph.D. Director of Advanced Development Variantyx, Inc. 1671 Worcester Rd Ste 300 Framingham, MA 0170 (617) 209-2090 <6172092090>

-- The contents of this message, together with any attachments, are intended only for the use of the person(s) to which they are addressed and may contain confidential and/or privileged information. If you are not the  intended recipient, immediately advise the sender and delete this message  and any attachments. Any distribution, or copying of this message, or any  attachment, is prohibited.

[vyx-lucy-kaplun]

@jrobinso I believe we were able to pinpoint the issue with 5mC: it is related to CRAM format. The 2-color system works as expected when BAM format is used, but does not work with the same files in CRAM.

[jrobinso]

I can reproduce the error by converting the BAM file you sent to CRAM.

[jrobinso]

@vyx-lucy-kaplun Until this is fixed you won't be able to use CRAM files to view base modifications. I will update this issue when we release a fix.

@cmdcolin It seems that the "ML" tag, or perhaps any tag of type "B,C", is not correctly parsed. For example, on an alignment in which I expect an array of size 656 for the ML tag I get an array of size 144. The values are correct as far as they go, at least the first few values are correct, its just truncated.

[vyx-lucy-kaplun]

@jrobinso Thank you!

[jrobinso]

This is fixed in release 3.0.7

[vyx-lucy-kaplun]

@jrobinso Thank you! It works now