search

The-Sequence-Ontology / SO-Ontologies

Collect of SO Ontologies
Creative Commons Attribution 4.0 International
96 stars 37 forks source link

Is there an appropriate term for reference allele? #334

Closed dandanxu closed 8 years ago

dandanxu commented 9 years ago

For when the sequenced allele is not different from the reference genome. See screen shot 2015-10-13 at 12 58 59. We currently have an effect predictor that outputs everything in sequence ontology terms, but right now, if the variant is the reference allele at the locus, we don't have a good way of displaying that "predicted effect" or term for this.

srynobio commented 9 years ago

In our collaborations with ClinGen we are currently working on adding a term to SO to handle this very thing.

The term being considered is:

id: no_sequence_alteration

Currently we are working on the final definition, but it would specify a sequence location where the observed allele matches the reference.

Any additional input from the community would be welcome.

thefferon commented 9 years ago

NCBI had the following exchange with SO in January 2014:

  1. With the emergence of personal genomes it is increasingly important to represent not only where a person's genome deviates from the reference, but to be explicit when a locus has been typed and specifically does NOT deviate from the reference. This should be reported as positive data (rather than remain unreported because it does not differ from the reference).

Request: Please add the term 'no_variation' to the SO:

term: no_variation definition: a sequence at a defined location that is defined to match the reference assembly synonyms: reference match parent: sequence_alteration This one is difficult to place in the ontology for obvious reasons - it is not an alteration. There has not previously been a use case for this as a variant file would not contain it. Can you explain more about your use case for this? The only solution to fit this into the ontology would be to have a parent class of sequence alteration and have the new term be a sibling of alteration. I don't know what the super class would be. We definitely need to talk thru this one…

I don’t believe it was ever followed up on or resolved.

On Oct 13, 2015, at 3:10 PM, Shawn Rynearson notifications@github.com<mailto:notifications@github.com> wrote:

In our collaborations with ClinGenhttps://www.clinicalgenome.org/ we are currently working on adding a term to SO to handle this very thing.

The term being considered is:

id: no_sequence_alteration

Currently we are working on the final definition, but it would specify a sequence location where the observed allele matches the reference.

Any additional input from the community would be welcome.

— Reply to this email directly or view it on GitHubhttps://github.com/The-Sequence-Ontology/SO-Ontologies/issues/334#issuecomment-147817355.

thefferon commented 9 years ago

– The fault likely being mine.

On Oct 13, 2015, at 3:20 PM, Tim Hefferon theffero@ncbi.nlm.nih.gov<mailto:theffero@ncbi.nlm.nih.gov> wrote:

NCBI had the following exchange with SO in January 2014:

  1. With the emergence of personal genomes it is increasingly important to represent not only where a person's genome deviates from the reference, but to be explicit when a locus has been typed and specifically does NOT deviate from the reference. This should be reported as positive data (rather than remain unreported because it does not differ from the reference).

Request: Please add the term 'no_variation' to the SO:

term: no_variation definition: a sequence at a defined location that is defined to match the reference assembly synonyms: reference match parent: sequence_alteration This one is difficult to place in the ontology for obvious reasons - it is not an alteration. There has not previously been a use case for this as a variant file would not contain it. Can you explain more about your use case for this? The only solution to fit this into the ontology would be to have a parent class of sequence alteration and have the new term be a sibling of alteration. I don't know what the super class would be. We definitely need to talk thru this one…

I don’t believe it was ever followed up on or resolved.

On Oct 13, 2015, at 3:10 PM, Shawn Rynearson notifications@github.com<mailto:notifications@github.com> wrote:

In our collaborations with ClinGenhttps://www.clinicalgenome.org/ we are currently working on adding a term to SO to handle this very thing.

The term being considered is:

id: no_sequence_alteration

Currently we are working on the final definition, but it would specify a sequence location where the observed allele matches the reference.

Any additional input from the community would be welcome.

— Reply to this email directly or view it on GitHubhttps://github.com/The-Sequence-Ontology/SO-Ontologies/issues/334#issuecomment-147817355.

dandanxu commented 9 years ago

@srynobio Great! We would be delighted to see something like no_sequence_alteration. Any ideas on when this might be implemented? Also, is there an open list/discussion somewhere on SO terms the Clingen/SO collaboration are planning on defining? We (at SolveBio) have been actively developing our variant-specific algorithms around SO terms and it'd be great to be able to know what's coming up.

srynobio commented 9 years ago

Coincidence, your solvebio webpage was just brought up today with one of our ClinGen calls, it look interesting.

As far as an open list/discussion we currently do not have one to my knowledge, the term no_sequence_alteration is being added to aid our Data Model efforts.

However, ideas for terms and improvements can always be submitted via github issue.

dandanxu commented 9 years ago

Oh how random! Let me know if you want a walkthrough anytime, we're always open to feedback and collaborations: dandan@solvebio.com. We will definitely submit issues/ideas for SO as we come across them.

keilbeck commented 8 years ago

We settled on this term no_variation SO:0002073 a sibling of sequence_alteration and a child of sequence_comparison

thefferon commented 8 years ago

Hi Karen,

NCBI inquired a while back about requesting an SO term for “no variation” – i.e., an assertion that an allele was probed and found to NOT be different from the reference.

Is no_variation SO:0002073 the answer to this? When was it / will it be added? I don’t find it in the current release or SVN in the SO Browser -

Thanks, Tim

On Mar 7, 2016, at 12:32 PM, Karen EIlbeck notifications@github.com<mailto:notifications@github.com> wrote:

We settled on this term no_variation SO:0002073 a sibling of sequence_alteration and a child of sequence_comparison

— Reply to this email directly or view it on GitHubhttps://github.com/The-Sequence-Ontology/SO-Ontologies/issues/334#issuecomment-193361803.

keilbeck commented 8 years ago

It was added in the fall. It should be there. I am going to save a revision today and make a release. —K

Karen Eilbeck Associate Professor Biomedical Informatics University of Utah

From: thefferon notifications@github.com<mailto:notifications@github.com> Reply-To: The-Sequence-Ontology/SO-Ontologies reply@reply.github.com<mailto:reply@reply.github.com> Date: Monday, March 7, 2016 at 10:39 AM To: The-Sequence-Ontology/SO-Ontologies SO-Ontologies@noreply.github.com<mailto:SO-Ontologies@noreply.github.com> Cc: Karen Eilbeck keilbeck@genetics.utah.edu<mailto:keilbeck@genetics.utah.edu> Subject: Re: [SO-Ontologies] Is there an appropriate term for reference allele? (#334)

Hi Karen,

NCBI inquired a while back about requesting an SO term for “no variation” – i.e., an assertion that an allele was probed and found to NOT be different from the reference.

Is no_variation SO:0002073 the answer to this? When was it / will it be added? I don’t find it in the current release or SVN in the SO Browser -

Thanks, Tim

On Mar 7, 2016, at 12:32 PM, Karen EIlbeck notifications@github.com<mailto:notifications@github.commailto:notifications@github.com> wrote:

We settled on this term no_variation SO:0002073 a sibling of sequence_alteration and a child of sequence_comparison

— Reply to this email directly or view it on GitHubhttps://github.com/The-Sequence-Ontology/SO-Ontologies/issues/334#issuecomment-193361803.

— Reply to this email directly or view it on GitHubhttps://github.com/The-Sequence-Ontology/SO-Ontologies/issues/334#issuecomment-193364203.

dandanxu commented 8 years ago

Does the sequence ontology website (http://www.sequenceontology.org/browser/obob.cgi) tend to be updated much later than the .obo files? I see the no_variation in the .obo files, but its not on the MISO web pages (which we mostly work off).

keilbeck commented 8 years ago

I think there must be a glitch somewhere. I will look into it. You are the second person today to comment about that.

--K

thefferon commented 8 years ago

Hi Karen,

Any progress on this? I still don't see 'no_variation SO:0002073' in the Miso browser. ("I don't see no variation???")

NCBI requests an update to this term's definition, please.

Current definition: "No_variation is a position or feature within a sequence, when compared to a reference sequence, that does not display any variation."

Proposed definitions (please comment on which is preferable):

1. "a position or feature within a sequence that is identical to the comparable position or feature of a specified reference sequence"

2. "a locus on a sequence that is identical to the comparable locus on a specified reference sequence"

–Tim

thefferon commented 8 years ago

Another question: Since no_alteration is a sibling of sequence_alteration, and the GVF spec states all terms must be sequence_alteration or a child of it, I conclude that no_variation would not be a valid term to specify in a GVF file.

If that is true, it is also problematic.

keilbeck commented 8 years ago

We need to alter the spec to be the parent of sequence alteration

Karen Eilbeck Associate Professor Biomedical Informatics University of Utah

From: thefferon notifications@github.com<mailto:notifications@github.com> Reply-To: The-Sequence-Ontology/SO-Ontologies reply@reply.github.com<mailto:reply@reply.github.com> Date: Tuesday, March 29, 2016 at 12:11 PM To: The-Sequence-Ontology/SO-Ontologies SO-Ontologies@noreply.github.com<mailto:SO-Ontologies@noreply.github.com> Cc: Karen Eilbeck keilbeck@genetics.utah.edu<mailto:keilbeck@genetics.utah.edu> Subject: Re: [The-Sequence-Ontology/SO-Ontologies] Is there an appropriate term for reference allele? (#334)

Another question: Since no_alteration is a sibling of sequence_alteration, and the GVF spec states all terms must be sequence_alteration or a child of it, I conclude that no_variation would not be a valid term to specify in a GVF file.

If that is true, it is also problematic.

— You are receiving this because you modified the open/close state. Reply to this email directly or view it on GitHubhttps://github.com/The-Sequence-Ontology/SO-Ontologies/issues/334#issuecomment-203032042

thefferon commented 8 years ago

@keilbeck , Any comment on the proposed values for redefining SO:0002073? Thanks

keilbeck commented 8 years ago

Hi TIm I can update the definition to the 1st one you gave: "a position or feature within a sequence that is identical to the comparable position or feature of a specified reference sequence"

Does that work for you still?

Shawn is working on getting miSO to work with git.

thefferon commented 8 years ago

Hi Karen, Great! Thanks for the follow-up. Is there a ticket for updating the GVF spec to "parent of sequence_alteration"? Should I open one? Thanks, Tim

keilbeck commented 8 years ago

Please open one. That would be great.

thefferon commented 8 years ago

@keilbeck – OK! Is there a separate github section for GVF, or shall I open a ticket in SO-Ontologies?

keilbeck commented 8 years ago

Put it in SO for now. THanks

thefferon commented 8 years ago

@dandanxu, @srynobio: Are there any persisting incompatibilities (especially w.r.t. ClinGen-related workflows) between the term SO:0002073 (no_variation) and the term discussed above (no_sequence_alteration)? If there is a compelling need for the term to be no_sequence_alteration, and if @keilbeck agrees to the possibility of a change, I can find out if NCBI would have any problems updating the term name. OR - could it be listed as an alias?

keilbeck commented 8 years ago

Hi all

I have moved sequence comparison to be the parent of sequence alteration. I have renamed no_variation to be no_sequence_alteration It is a sibling of sequence_alteration. I updated the definition: A position or feature within a sequence that is identical to the comparable position or feature of a specified reference sequence.

Tim, do you need to go in and update the spec?

Does this fix your immediate issues? --K

keilbeck commented 8 years ago

It is checked in now Tim, look for the latest revision.

dandanxu commented 8 years ago

@thefferon my apologies didn't see this thread till now. we have no problem with no_sequence_alteration. looking forward to seeing it on MISO (is it a new SO id or still SO:0002073?).

srynobio commented 8 years ago

I have no issue as well, originally no_sequence_alteration was the term I prefered, I also prefer the new placement within the ontology.

thefferon commented 8 years ago

@dandanxu ,

is it a new SO id or still SO:0002073?

SO:0002073: no_sequence_alteration is a sibling of SO:0001059: sequence_alteration both the above are children of SO:0002072: sequence_comparison which is a child of SO:0000110: sequence_feature