Unable to specify sex of individual and tryvamos does not parse vcf correctly

[chrisclarkson]

Hello
I am trying to use vamos to visualise a waterfall plot for a single male.
My locus of interest is on the X chromosome.
There does not seem to be any obvious way to specify the individual's sex-
vamos --read -h does not mention anything about sex/karyotype.
I really need to as it is important for my analysis- could this be made possible?

Furthermore I am confused when trying to do a waterfall plot:
I input the vcf to tryvamos.py
but on line ~79 it returns the error:

end,ru,_,annos = info.split(';')
More values to unpack than expected.

I inspected info and it contains 6-7 ';' marks:

END=....;RU=CAG,CCA,SVTYPE=VNTR,ALT_ANNO_H1=0000.....;LEN_H1=...;ALT_ANNO_H1=0000.....;LEN_H1=...;

I would manually edit the code manually but I am not sure what needs to be done with annos after it is created- hence I'm not sure how to proceed.
Can you help with this- also it would be really useful if I could specify the sex.

[BidaGu]

Hi,

Since waterfall plot is on a panel of samples, the "waterfallPlot" option of tryvamos works on the combined multi-sample vcf generated from the "combineVCF" option. The INFO field of a single sample vcf contains more than 4 fields, while the INFO field of a combined multi-sample vcf contains 4 fields as the code suggests.

Alternatively, if you want to visualize the motif annotation of a single sample, you can generate a "combined" vcf using this single sample entry and then run the "waterfallPlot" command.

[mchaisso]

I can add the code to specify sex, but I'm not super available for a few weeks. As a stopgap, the phasing algorithm should estimate all male chrX reads to be autozygous, and emit identical genotypes for each chromosome. Any regions this is not the case will reflect collapsed duplications or other unreliable alignments. Are you able to check this?

[chrisclarkson]

Hi there thank you for your quick response.
Processing the single vcf with combineVCF worked and I can now plot the result.

I should have mentioned that this is an STR- my bed file is formatted as follows-example:
chrX 100000 100030 CAG,CAA v-2.0 STR 3 0 0

My command is:
vamos --read -b lr.bam -r in.bed -s sample -o out.vcf -t 8
Firstly the output VCF says VNTR and STR does not appear anywhere... is that a problem? If so why is the STR label that I have provided not being recognised? Am I formatting it wrong?
Here is the docker image I pulled to get this working:
quay.io/biocontainers/vamos:2.1.5--h5ca1c30_1
As for plotting the data- it definitely recognises 2 different alleles in the data- see below:

Does this give the answer you need to your second comment?

Furthermore- out of interest I also tried the vamos somatic option which returns alleles containing 000000111111000001111110000....
This is much more what I was expecting to see as my read data seem to have large CAA interruptions in the middle of the CAG tracks. And yet the --read option only returns alleles that show single interruptions: 00001000100...

Is there something that I am doing to make this happen?
Again thanks for your help just looking to get final bit of help with this.

[chrisclarkson]

Sorry in response to my comment about interruptions- it turns out that I was making a mistake in specifying the coordinates- I was setting them according to the tandem repeat locus in the hg38 genome. If I widen them- then your software returns something a bit more similar to what I was expecting- however it still returns fairly different allele sizes for the separate alleles: ~1500 in A1 and ~2100 in A2- is this still in line with what you expect?

Do you recommend a particular locus width? Currently I can't plot the results as +/-1000bp is too wide for the python waterfall script to handle.

Also I still can't get it to recognise the locus as an STR rather than VNTR.

[mchaisso]

The output currently always states VNTR (STR support was only recently added). We can post a fix to keep track of the bed file annotation, but for now,VNTR applies to STR/VNTR. The region that you are looking at is a segmental duplication: https://genome.ucsc.edu/cgi-bin/hgTracks?db=hg38&lastVirtModeType=default&lastVirtModeExtraState=&virtModeType=default&virtMode=0&nonVirtPosition=&position=chrX%3A85016%2D115015&hgsid=2566458548_AF70Kn9v3P1m4zr6AJ8doU7nCQek These are hotspots for extra duplication, so it is likely your sample has an extra copy of this sequence relative to hg38, and the phasing algorithm treats it as a separate chromosome. If you look at the IGV views of the bam, does it show two separate patterns of insertion?

…

On Fri, May 16, 2025 at 5:25 AM Chris Clarkson ***@***.***> wrote: *chrisclarkson* left a comment (ChaissonLab/vamos#21) <#21 (comment)> Sorry in response to my comment about interruptions- it turns out that I was making a mistake in specifying the coordinates- I was setting them according to the tandem repeat locus in the hg38 genome. If I widen them- then your software returns something a bit more similar to what I was expecting- however it still returns fairly different allele sizes for the separate alleles: ~1500 in A1 and ~2100 in A2- is this still in line with what you expect? Also I still can't get it to recognise the locus as an STR rather than VNTR. — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKPGPW6EWZUOONY4LUPANL26XKMTAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQOBWGU3TONBUGE> . You are receiving this because you commented.Message ID: ***@***.***>

[chrisclarkson]

sorry that region that I gave to you was just an example locus -if you need the real one it is here:

chrX:19990919-19990973 motif=CCG
Still having the same problem with it calling 2 very different repeat sizes- I can't give you the bam file unfortunately as it is behind a closed wall.

[mchaisso]

Ok, that is an error in the phasing then. I'll track down a male sample here to see what is happening. Is this ONT or PacBio?

…

On Mon, May 19, 2025 at 8:06 AM Chris Clarkson ***@***.***> wrote: *chrisclarkson* left a comment (ChaissonLab/vamos#21) <#21 (comment)> sorry that region that I gave to you was just an example locus -if you really need the real one it is here: chrX:19990919-19990973 motif=CCG Still having the same problem with it calling 2 very different repeat sizes- I can't give you the bam file unfortunately as it is behind a closed wall. — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKPGPUAUGKFVBG3QRXZODD27HXQXAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQOJRGM3TONRQGE> . You are receiving this because you commented.Message ID: ***@***.***>

[chrisclarkson]

ONT- thank you for your help on this

[mchaisso]

Just giving an update. I couldn't replicate the error on HG002, a male. There was a separate problem of double-initialization of the input chromosomes that is fixed in version 2.1.6 (in git and pushed to docker), but I do not anticipate this caused the problem. The STR/VNTR annotation in the vcf is fixed. One potential issue is that the locus is near the start of the chromosome, and some reads may have strange alignments at their ends. Can you try adding -p 1000 as a workaround? Thanks, -Mark

…

On Mon, May 19, 2025 at 9:16 AM Chris Clarkson ***@***.***> wrote: *chrisclarkson* left a comment (ChaissonLab/vamos#21) <#21 (comment)> ONT- thank you for your help on this — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKPGPXJ2LG4N2AHLXUCP7L27H7VDAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQOJRGU4DKMZSGQ> . You are receiving this because you commented.Message ID: ***@***.***>

[chrisclarkson]

Hiya
I tried with -p 1000 it gives GT 1/2 and H1_len=83 and H2_len=330.
It could be down to the fact that this repeat allele is expanded....?

Kind Regards
Chris

[mchaisso]

The only way two alleles can be output here is if there are two partitions of reads by phase. Is it possible to continue the conversation over e-mail?

…

On Fri, May 30, 2025 at 4:48 PM Chris Clarkson ***@***.***> wrote: *chrisclarkson* left a comment (ChaissonLab/vamos#21) <#21 (comment)> Hiya I tried with -p 1000 it gives GT 1/2 and H1_len=83 and H2_len=330. It could be down to the fact that this repeat allele is expanded....? Kind Regards Chris — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKPGPXKUB5DDXY4IFDX6LL3BDU5HAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMRTG42DIMRZGM> . You are receiving this because you commented.Message ID: ***@***.***>

[chrisclarkson]

Hello thanks Yes no problem Here is my email- ***@***.*** Kind regards Chris Sent from Outlook for iOS<https://aka.ms/o0ukef>

________________________________ From: Mark Chaisson ***@***.***> Sent: Saturday, May 31, 2025 11:30:30 PM To: ChaissonLab/vamos ***@***.***> Cc: Clarkson, Chris ***@***.***>; Author ***@***.***> Subject: Re: [ChaissonLab/vamos] Unable to specify sex of individual and tryvamos does not parse vcf correctly (Issue #21) [https://avatars.githubusercontent.com/u/1372990?s=20&v=4]mchaisso left a comment (ChaissonLab/vamos#21)<#21 (comment)> The only way two alleles can be output here is if there are two partitions of reads by phase. Is it possible to continue the conversation over e-mail?

On Fri, May 30, 2025 at 4:48 PM Chris Clarkson ***@***.***> wrote: *chrisclarkson* left a comment (ChaissonLab/vamos#21) <#21 (comment)> Hiya I tried with -p 1000 it gives GT 1/2 and H1_len=83 and H2_len=330. It could be down to the fact that this repeat allele is expanded....? Kind Regards Chris — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKPGPXKUB5DDXY4IFDX6LL3BDU5HAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMRTG42DIMRZGM> . You are receiving this because you commented.Message ID: ***@***.***>

— Reply to this email directly, view it on GitHub<#21 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AGMIVP3D2PGPAD5D2F2NIHD3BIUQNAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMRVHA4DIMRXG4>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[mchaisso]

Git hides e-mails. I'm mchaisso at usc dot edu. Thanks, -Mark On Sat, May 31, 2025 at 3:33 PM Chris Clarkson ***@***.***> wrote:

…

*chrisclarkson* left a comment (ChaissonLab/vamos#21) <#21 (comment)> Hello thanks Yes no problem Here is my email- ***@***.*** Kind regards Chris Sent from Outlook for iOS<https://aka.ms/o0ukef> ________________________________ From: Mark Chaisson ***@***.***> Sent: Saturday, May 31, 2025 11:30:30 PM To: ChaissonLab/vamos ***@***.***> Cc: Clarkson, Chris ***@***.***>; Author ***@***.***> Subject: Re: [ChaissonLab/vamos] Unable to specify sex of individual and tryvamos does not parse vcf correctly (Issue #21) [https://avatars.githubusercontent.com/u/1372990?s=20&v=4]mchaisso left a comment (ChaissonLab/vamos#21)< #21 (comment)> The only way two alleles can be output here is if there are two partitions of reads by phase. Is it possible to continue the conversation over e-mail? On Fri, May 30, 2025 at 4:48 PM Chris Clarkson ***@***.***> wrote: > *chrisclarkson* left a comment (ChaissonLab/vamos#21) > <#21 (comment)> > > Hiya > I tried with -p 1000 it gives GT 1/2 and H1_len=83 and H2_len=330. > It could be down to the fact that this repeat allele is expanded....? > > Kind Regards > Chris > > — > Reply to this email directly, view it on GitHub > <#21 (comment)>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AAKPGPXKUB5DDXY4IFDX6LL3BDU5HAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMRTG42DIMRZGM> > . > You are receiving this because you commented.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub< #21 (comment)>, or unsubscribe< https://github.com/notifications/unsubscribe-auth/AGMIVP3D2PGPAD5D2F2NIHD3BIUQNAVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMRVHA4DIMRXG4>. You are receiving this because you authored the thread.Message ID: ***@***.***> — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKPGPSF74V5ZLD6PTGPMG33BIU37AVCNFSM6AAAAAB5GFD2X6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMRVHA4DMNBSGU> . You are receiving this because you commented.Message ID: ***@***.***>