Decide on relationship between renal column and renal medulla

[dosumis]

The ASCT+B kidney table (1.1 beta) relates renal column and renal medulla to each other via ccf_part_of. Validation against Uberon shows that there is currently no is_a, part_of or overlaps relationship between 'renal column - UBERON_0001284' and 'renal medulla - UBERON_0000362'

Report (showing names used in table):

n	s	slabel	user_slabel	o	olabel	user_olabel
1	UBERON:0001284	renal column (column of Bertin)	renal column	UBERON:0000362	renal medulla	renal medulla

Uberon has 'renal column' 'part of' some 'renal cortex' but also has the textual definition: "A column of tissue that is histologically identical to tissue found in the renal cortex, but which is located in the medulla between the renal pyramids."

@emquardokus - who should we be speaking to at KPMG about this?

[emquardokus]

Kidney Precision Medicine Project (KPMP)Todd Valerius MTVALERIUS@bwh.harvard.edu, Sanjay Jain sanjayjain@wustl.edu and possibly Oliver He yongqunh@med.umich.edu

[paolaroncaglia]

@dosumis @emquardokus

1. 'renal column' has an editor note "Check part_of location against textual definition"
2. "The functional substance, or parenchyma, of the kidney is divided into two major structures: the outer renal cortex and the inner renal medulla. Grossly, these structures take the shape of eight to 18 cone-shaped renal lobes, each containing renal cortex surrounding a portion of medulla called a renal pyramid.[13] Between the renal pyramids are projections of cortex called renal columns." (https://en.wikipedia.org/w/index.php?title=Kidney&oldid=1050157496#Gross_anatomy)
3. "The renal column (or Bertin column, or column of Bertin) is a medullary extension of the renal cortex in between the renal pyramids. It allows the cortex to be better anchored." (https://en.wikipedia.org/w/index.php?title=Renal_column&oldid=963928030) => I'd suggest 'renal column' has_part 'renal cortex', or even 'renal column' is_a 'renal cortex'; 'renal column' adjacent_to 'kidney pyramid', or even 'renal column' surrounds 'kidney pyramid'?

We'd need to open a Uberon ticket to document suggestions and discussion (if any). I can open the ticket and then maybe Ellen can tag the experts there please?

[dosumis]

We'd need to open a Uberon ticket to document suggestions and discussion (if any). I can open the ticket and then maybe Ellen can tag the experts there please?

Yes please. In general I think we should be opening these as Uberon tickets rather than CCF in order to maximise input (my bad in this case). @emquardokus - in this case you should be able to tag it experts using GitHub handles I think.

[paolaroncaglia]

Please see my suggestions in https://github.com/obophenotype/uberon/issues/2126#issuecomment-945863982. @emquardokus , let me know if I can help any further at this point. Thanks.

[paolaroncaglia]

Hi @dosumis and @emquardokus , Following up on this please. I think Ellen mentioned that she'd be meeting the kidney experts quoted above. Please let me know if this issue was resolved, or if we should follow up with the experts, or ask other Uberon contributors for their feedback. Thanks.

[dosumis]

@tvalerius - would you be able to comment on this issue?

[paolaroncaglia]

Hi @dosumis @emquardokus, I know that you'd like to address this in time for inclusion in a bioRxiv submission. Was there any feedback from experts offline that I could incorporate? If not, would you like me to make a pull request and/or discuss at the Uberon call next Monday?

[dosumis]

@emquardokus - any progress on this? I'm just discussing this with Paola. One way to resolve this would be to ask reviewers whether they expect a query for cell types in the medulla to return cells in renal columns. If the answer is yes, the we can use 'part of', of not, we should consider another relationship (and maybe question whether the table is correct as it is meant to imply part_of ?)

[RDruzinsky]

I believe it is more proper to make the renal columns 'part of' the cortex rather than the medulla. The columns are the borders of the renal lobes, but https://doi.org/10.1016/0301-5629(86)90068-2 "cortical tissue in between neighbouring pyramids."

describes them as Robert E. Druzinsky, Ph.D. Clinical Associate Professor Dept. of Oral Biology College of Dentistry University of Illinois at Chicago 801 S. Paulina Chicago, IL 60612 @.***

Office: 312-996-0406 Lab: 312-996-0629 Website: www.peerj.com/RobertDruzinsky

On Tue, Jan 18, 2022 at 8:12 AM David Osumi-Sutherland < @.***> wrote:

@emquardokus https://github.com/emquardokus - any progress on this? I'm just discussing this with Paola. One way to resolve this would be to ask reviewers whether they expect a query for cell types in the medulla to return cells in renal columns. If the answer is yes, the we can use 'part of', of not, we should consider another relationship (and maybe question whether the table is correct as it is meant to imply part_of ?)

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/uberon/issues/2126#issuecomment-1015448996, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABASAOGZNYFNMN2BT42E4PTUWVYKZANCNFSM5GUUAHNA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you are subscribed to this thread.Message ID: @.***>

[emquardokus]

@dosumis @paolaroncaglia sent email to Sanjay Jain (HuBMAP & KPMP) and M. Todd Valerius (KPMP & GUDMAP) regarding this; however, I agree with R Druzinsky.

Based on the definition that I could find of renal column, it’s part of the cortex that dips into the medullary area between the renal pyramids, so cortex and not medulla, correct?

Would any cell types be present in both locations?

"The renal column (or Bertin column, or column of Bertin) is a medullary extension of the renal cortex in between the renal pyramids. It allows the cortex to be better anchored. Each column consists of lines of blood vessels and urinary tubes and a fibrous material."

The ASCT+B tables assume that the relationships between anatomical structures are a part to whole relationship (part_of).

Currently in the kidney table, when sending this through the Uberon validation program (CCF_Tools on GitHub) that David Osumi-Sutherland’s group and Katy’s group have developed, the renal column is AS/3 and renal medulla is AS/2 meaning that the implied relationship between them is renal column is part of renal medulla. This relationship is not currently in Uberon. (see screenshot of ASCT+B kidney v1.1)

[paolaroncaglia]

@RDruzinsky @emquardokus thanks for your feedback. So the consensus for now is that renal column should be part of the renal cortex. This is in fact already in place in Uberon: 'renal column' 'part of' some 'outer cortex of kidney' (and 'outer cortex of kidney' is subclass of 'cortex of kidney'). So unless we're missing something, the ontology seems to be correct, while the table may contain an inconsistency. @emquardokus , unless the experts you emailed object to the current ontology representation (or unless they've replied to you offline), I'll close the Uberon ticket, and you may address this in ASCT+B instead. Thanks.

[uberon]

This may be an opportunity for more precise spatial axioms, e.g overlaps vs part-disjoint-from

Message ID: @.***>

[paolaroncaglia]

@cmungall

This may be an opportunity for more precise spatial axioms, e.g overlaps vs part-disjoint-from …

I'm not sure what axiom exactly you're suggesting. Would that be between 'renal column' and 'outer cortex of kidney'? If the first is basically made of the same stuff than the second, how would 'overlap' apply? Honest question :-) there are so many spatial axioms I've never used.

[paolaroncaglia]

@dosumis @emquardokus we haven't heard back from the experts yet, and I think that this time Uberon's correct, can we close this ticket?

[uberon]

close it!

On Tue, Jan 25, 2022 at 3:50 AM paolaroncaglia @.***> wrote:

@dosumis https://github.com/dosumis @emquardokus https://github.com/emquardokus we haven't heard back from the experts yet, and I think that this time Uberon's correct, can we close this ticket?

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/uberon/issues/2126#issuecomment-1021107279, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAGZRCBXP3SUROMLH7G2XA3UX2FB7ANCNFSM5GUUAHNA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[dosumis]

Seems like the ontology has spoken and wishes this ticket to be closed. :)

[emquardokus]

@paolaroncaglia @dosumis Reply from Kidney Precision Medicine Project expert Dr. Sanjay Jain: Hi Ellen Sorry for the delay, I found your email in my spam. Yes this is a tricky one and probably have folks on both sides of the aisle. The definition in Uberon states it is part of medulla https://www.ebi.ac.uk/ols/ontologies/uberon/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FUBERON_0001284&viewMode=PreferredRoots&siblings=false while the anatomical tree is showing as part of cortex because it is extension of outer cortex diving deep into the kidney all the way to the pelvis. Histologically, this is pretty similar to cortical tissue so many for that reason may call it cortex. But traditionally we mark the kidney in 2 regions, cortex (outer) and medulla (inner part). So anatomically this would be in the inner part between the medullary pyramids. But then we have interlobar vessels that run between the medullary pyramid and the renal columns so what should we call them to be anatomically a part of. And then Uberon defines renal medulla to be synonymous with renal pyramids. Most people talking about renal medulla actually are really implying renal pyramids, and when talking about cortex, it is the cortical arch that is being referred to. So this comes to your question, if we say cortex what cell types will we expect to see there, and same for medulla. So how do we reconcile these conflicting issues of regional anatomy and histology and common usage. It is sort of what we struggled with separating renal corpuscle from glomerulus. One could be more precise and say cortical arch, medullary pyramid, renal column as anatomical references in the human kidney. Any way we can discuss what the ontologists suggest for consistency and see if the community would buy in. Thanks Sanjay

[RDruzinsky]

I think that the renal columns should be 'part of' some renal cortex.

They are extensions of the renal cortex.

"Prominent Columns of Bertin Prominent (or hypertrophied) columns of Bertin (also known as “septa of Bertin”) is a normal variant of the kidney formed of hyper-trophied cortical tissue located between the pyramids that projects into the renal sinus, resulting in splaying of the sinus. Hypertrophied but otherwise normal renal tissue not only distorts the renal sinus but can also result in what appear to be bizarre or disorganized calyces on excretory urography or CT urography because the papillae are not situated in the typical radial orientation." ( https://www.ajronline.org/doi/full/10.2214/AJR.06.0920)

Robert E. Druzinsky, Ph.D. Clinical Associate Professor Dept. of Oral Biology College of Dentistry University of Illinois at Chicago 801 S. Paulina Chicago, IL 60612 @.***

Office: 312-996-0406 Lab: 312-996-0629 Website: www.peerj.com/RobertDruzinsky

On Wed, Jan 26, 2022 at 9:48 AM emquardokus @.***> wrote:

Reply from Kidney Precision Medicine Project expert Dr. Sanjay Jain: Begin forwarded message: From: "Jain, Sanjay" @. Subject: RE: HuBMAP: Question regarding relationship between renal column and renal medulla Date: January 26, 2022 at 12:03:57 AM EST To: "Quardokus, Ellen M" @., "M. Todd Valerius" @. Cc: David Osumi-Sutherland @., Roncaglia Paola @.*** Hi Ellen Sorry for the delay, I found your email in my spam. Yes this is a tricky one and probably have folks on both sides of the aisle. The definition in Uberon states it is part of medulla https://www.ebi.ac.uk/ols/ontologies/uberon/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FUBERON_0001284&viewMode=PreferredRoots&siblings=false while the anatomical tree is showing as part of cortex because it is extension of outer cortex diving deep into the kidney all the way to the pelvis. Histologically, this is pretty similar to cortical tissue so many for that reason may call it cortex. But traditionally we mark the kidney in 2 regions, cortex (outer) and medulla (inner part). So anatomically this would be in the inner part between the medullary pyramids. But then we have interlobar vessels that run between the medullary pyramid and the renal columns so what should we call them to be anatomically a part of. And then Uberon defines renal medulla to be synonymous with renal pyramids. Most people talking about renal medulla actually are really implying renal pyramids, and when talking about cortex, it is the cortical arch that is being referred to. So this comes to your question, if we say cortex what cell types will we expect to see there, and same for medulla. So how do we reconcile these conflicting issues of regional anatomy and histology and common usage. It is sort of what we struggled with separating renal corpuscle from glomerulus. One could be more precise and say cortical arch, medullary pyramid, renal column as anatomical references in the human kidney. Any way we can discuss what the ontologists suggest for consistency and see if the community would buy in. Thanks Sanjay

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/uberon/issues/2126#issuecomment-1022329945, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABASAODNEASEBETZO4LS7K3UYAJURANCNFSM5GUUAHNA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[paolaroncaglia]

@emquardokus @RDruzinsky cc @dosumis thank you. I agree with Robert that renal columns should (continue to) be represented in Uberon as part of renal cortex. OTOH I appreciate that there may be "folks on both sides of the aisle". The only options that come to my mind to reconcile two opposite views (renal columns are/aren't part of renal cortex) in the ontology are: a) make renal columns part of kidney more broadly; add a comment to say that disagreements in the community make it hard to pick a more granular location; (please don't shoot the messenger, in fact this is not the option I'd favor personally) b) leave renal column as part of renal cortex; add a comment to mention that some members of the community may not expect this placement, but that we believe it to be correct based on e.g. the 2 references that Robert provided; c) make renal column part of renal medulla instead, but I believe that we haven't collected any published reference in support of this; d) before doing any of the above, discuss more extensively with experts/community members to see what option would meet the largest consensus. IMO, a) might be an easy fix, but some of us would find it unsatisfying; d) would require the longest, but might prevent future discussions. @dosumis @emquardokus please let me know what you'd prefer, based on the urgency (or not) of addressing this.

[RDruzinsky]

I think choice 3 is the most reasonable. I believe that Uberon should follow the biology as closely as possible, not opinion based on the history of the terminology. The renal columns have been thought of as part of the medulla only because they are in the inner, medullary region rather than the outer, cortical region. In reality, they are extensions of the cortex.

Robert E. Druzinsky, Ph.D. Clinical Associate Professor Dept. of Oral Biology College of Dentistry University of Illinois at Chicago 801 S. Paulina Chicago, IL 60612 @.***

Office: 312-996-0406 Lab: 312-996-0629 Website: www.peerj.com/RobertDruzinsky

On Wed, Jan 26, 2022 at 12:06 PM paolaroncaglia @.***> wrote:

@emquardokus https://github.com/emquardokus @RDruzinsky https://github.com/RDruzinsky cc @dosumis https://github.com/dosumis thank you. I agree with Robert that renal columns should (continue to) be represented in Uberon as part of renal cortex. OTOH I appreciate that there may be "folks on both sides of the aisle". The only options that come to my mind to reconcile two opposite views (renal columns are/aren't part of renal cortex) in the ontology are: a) make renal columns part of kidney more broadly; add a comment to say that disagreements in the community make it hard to pick a more granular location; (please don't shoot the messenger, in fact this is not the option I'd favor personally) b) leave renal column as part of renal cortex; add a comment to mention that some members of the community may not expect this placement, but that we believe it to be correct based on e.g. the 2 references that Robert provided; c) make renal column part of renal medulla instead, but I believe that we haven't collected any published reference in support of this; d) before doing any of the above, discuss more extensively with experts/community members to see what option would meet the largest consensus. IMO, a) might be an easy fix, but some of us would find it unsatisfying; d) would require the longest, but might prevent future discussions. @dosumis https://github.com/dosumis @emquardokus https://github.com/emquardokus please let me know what you'd prefer, based on the urgency (or not) of addressing this.

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/uberon/issues/2126#issuecomment-1022456571, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABASAOGL7YACRC7ZU5ZE4B3UYAZ3NANCNFSM5GUUAHNA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[dosumis]

The renal columns have been thought of as part of the medulla only because they are in the inner, medullary region rather than the outer, cortical region.

Potential compromise: We make an additional term 'medullary region of kidney' which covers the pyramids and the columns?

[RDruzinsky]

@dosumis https://github.com/dosumis, if you think that would be a useful approach, I have no objection.

Robert E. Druzinsky, Ph.D. Clinical Associate Professor Dept. of Oral Biology College of Dentistry University of Illinois at Chicago 801 S. Paulina Chicago, IL 60612 @.***

Office: 312-996-0406 Lab: 312-996-0629 Website: www.peerj.com/RobertDruzinsky

On Wed, Jan 26, 2022 at 4:21 PM David Osumi-Sutherland < @.***> wrote:

The renal columns have been thought of as part of the medulla only because they are in the inner, medullary region rather than the outer, cortical region.

Potential compromise: We make an additional term 'medullary region of kidney' which covers the pyramids and the columns?

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/uberon/issues/2126#issuecomment-1022657324, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABASAOGL4SCHC652I56S4R3UYBXXTANCNFSM5GUUAHNA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[paolaroncaglia]

To start with, I created UBERON:8410073 'medullary region of kidney', see https://github.com/obophenotype/uberon/pull/2279/files. Next, I'll address 'renal column'.

[paolaroncaglia]

I edited 'renal column' to make it part of 'medullary region of kidney', modified definition and added comment, and revised/added def dbxrefs (the old one pointed to a link that's no longer active), see https://github.com/obophenotype/uberon/pull/2280/files.

[paolaroncaglia]

@dosumis @emquardokus @RDruzinsky Hopefully this is fixed now. Let me know if you have additional suggestions for 'medullary region of kidney' and/or 'renal column'. Links to the pull requests (with details of the edits) are in my 2 previous comments. If all is well as is, please close this ticket. Thanks.

[RDruzinsky]

Can we add something to medullary region that states that it is not an exact synonym of renal medulla?

On Thu, Feb 3, 2022, 9:03 AM paolaroncaglia @.***> wrote:

@dosumis https://github.com/dosumis @emquardokus https://github.com/emquardokus @RDruzinsky https://github.com/RDruzinsky Hopefully this is fixed now. Let me know if you have additional suggestions for 'medullary region of kidney' and/or 'renal column'. Links to the pull requests (with details of the edits) are in my 2 previous comments. If all is well as is, please close this ticket. Thanks.

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/uberon/issues/2126#issuecomment-1029080203, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABASAOCDFK25EZF6HRFO4WDUZKKK5ANCNFSM5GUUAHNA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[paolaroncaglia]

@RDruzinsky

Can we add something to medullary region that states that it is not an exact synonym of renal medulla?

Of course, thank you for suggesting that. I added rdfs:comments to both terms:

UBERON:0000362 renal medulla comment: Note that 'renal medulla' is not synonymous with 'medullary region of kidney'. The former is composed of kidney pyramids only, while the latter covers kidney pyramids and renal columns, which are histologically different.

UBERON:8410073 medullary region of kidney comment: Note that 'medullary region of kidney' is not synonymous with 'renal medulla'. The former covers kidney pyramids and renal columns, which are histologically different, while the latter is composed of kidney pyramids only.

Please let me know if there's anything else. Best,

Paola

[paolaroncaglia]

@dosumis @emquardokus cc @RDruzinsky I'll close this ticket now. Not sure if anything needs to be done on the ASCT+B kidney table side. If any further edit is required in Uberon, please open a new ticket. Thanks!