Open dosumis opened 3 years ago
Some general notes:
A. Lots of general tissue types as subjects in relationships where an organ specific term is needed for part_of or overlaps to apply: 'connective tissue'; 'nerve', 'adipose tissue'.
B. Lots of vasculature/lymphatic vessels -
C. 5 cases that need investigation (numbered here for easy reference)
The log table should be read as:
In Uberon+CL s
has no inferred or asserted is_a/part_of/overlaps relationship to o
.
n | s | slabel | user_slabel | o | olabel | user_olabel |
---|---|---|---|---|---|---|
1 | UBERON:0002125 | thymus lobule | thymus lobule | UBERON:0005483 | thymus lobe | thymus lobe |
2 | UBERON:0006936 | thymus subcapsular epithelium | thymus subcapsular epithelium | UBERON:0002125 | thymus lobule | thymus lobule |
3 | UBERON:0003846 | thymus lymphoid tissue | thymus lymphoid tissue | UBERON:0006936 | thymus subcapsular epithelium | thymus subcapsular epithelium |
4 | UBERON:0003483 | thymus epithelium | thymus epithelium | UBERON:0006936 | thymus subcapsular epithelium | thymus subcapsular epithelium |
5 | UBERON:0003987 | Hassall's corpuscle | Hassall's corpuscle | UBERON:0003483 | thymus epithelium | thymus epithelium |
The following numbered points refer to n
numbers in the table:
thymus lobule
has taxon_notes "in the mouse, the lobes of the thymus are not subdivided into lobules but only in the central medulla and a peripheral cortex (i.e. there is no distinct sublobulation). In humans, the two thymus lobes are composed of many lobules of various sizes which contain follicles, each comprising a medulla and a cortex. in the rat, the thymus is partially subdivided into lobules separated by thin bands of connective tissue which are continuous with the thin connective tissue capsule.[MP]".
@dosumis
Please see google sheet here where authors provided their viewpoint on New term requests with citations, superclass, and all the info needed to add; I had shared this yesterday: NewTermRequests_uberon_CL_ASCT+B https://docs.google.com/spreadsheets/d/1BihrOU9vWrsA_wYjcbdPUr42_Sb22CO3gkBDieiYQPs/edit?usp=sharing
Issue obophenotype/ccf-validation-tools#1 thymus lobules as part of lobes. YES, lobules are always part of lobes, but presence of lobules are taxon specific (not in mouse). This paper mentions/documents taxon differences between human and rodent and also some differences between rat, mouse, guinea pig for thymus lobes/lobules. Consensus is 2 lobes, separated by a septum of connective tissue, but formation of lobules differs.
In addition, we already also know there is a huge difference between child, adolescent, adult thymus tissue progression. Adult thymus is largely fat (adipose) tissue and dramatically shrinks in size from child to adult.
Differences exist between location of thymus between humans (thoracic cavity near heart) compared to rodents (cervical/neck) Pearse, Gail. 2006. “Normal Structure, Function and Histology of the Thymus.” Toxicologic Pathology 34 (5): 504–14. https://doi.org/10.1080/01926230600865549.
Issue obophenotype/ccf-validation-tools#2 The same reference I cited in Issue obophenotype/ccf-validation-tools#1 above goes part by part for thymus so is especially rich in information with respect to capsule and epithelium; it states: "The bulk of the supporting framework in the thymus is composed of the network of epithelial reticular cells. Epithelium-free areas (or “holes”) are compartments lacking stromal elements, which have been identified in the sub-capsular area, extending deep into the cortex (Bruijntjes et al., 1993; Elmore, 2006)." "Immunohistochemically, epithelial cells can be divided into four distinct subtypes: subcapsular cortical, inner cortical, medullary and Hassalls corpuscles. A decrease in the number of different antigenic epithelial populations occurs with age in mice. "
Issue 3 Relationship between lymphoid tissue and epithelium does appear to be special in thymus;I found this in literature for thymus: "Differentiating T-cells accumulate between epithelial reticular cells. They can then pass into the venules and efferent lymphatics along the border between the cortex and the medulla."
Issue 4 Agree about thymus epithelium <-- thymus subcapsular epithelium; however, I am wondering why the uberon relationship of thymus subcapsular epithelium and thymus capsule is not in table---instead thymus subcapsular epithelium is part of lobule instead, which doesn't make sense to me--did that former relationship "pass" the validation test?
Issue 5** I don't understand what this is referring to ? Is this referring to issue 3? See relationship described in paper quoted. "thymus lymphoid tissue" is part_of thymus, but I do not see most of implied relationships from ASCT+B table.
Issue 5 I don't understand what this is referring to ? s this referring to issue 3?
row n = 5 in the log table. Looks like a typo.
To avoid any confusion, I've added some clarification to my general notes comment on how to read the table and comments. Part of my aim here is to come up with a standard template/SOP for reviewing and actioning. One way to do that might be to automate the generation of tickets via the GitHub API - these can be versioned and dated and include boilerplate text describing how to read the table and document the issues. Feedback on how to improve tables and standardise comments would help. This is also the first time we've really tested this software in any big way. It's entirely possible that there are bugs, so any cases that don't make sense given the original table should be reported as a ticket on this repo and assigned to @anitacaron
As much as I enjoy it, I will have limited time to debate fine details of the biology. Where those impinge on decisions about modifying ontologies, could you discuss with @paolaroncaglia first, and pull me in or bring up on a call if more discussion is needed?
Where there is a clear modification - e.g. 'add 'thymus lobule' part_of some 'thymus lobe'
- please make an Uberon ticket and assign it to @paolaroncaglia (or try the edit yourself). Make sure to link the ticket from the review ticket on this repo.
FYI, I opened a CL ticket for the thymus new cell type term requests: https://github.com/obophenotype/cell-ontology/issues/1279
@emquardokus
Issue 5 is pointing to a general error in the Thymus table that needs fixing
thymus epithelium | thymus epithelium | UBERON:0003483 | medullary thymic epithelial cell type 1 | |||
---|---|---|---|---|---|---|
thymus epithelium | thymus epithelium | UBERON:0003483 | medullary thymic epithelial cell type 2 | |||
thymus epithelium | thymus epithelium | UBERON:0003483 | medullary thymic epithelial cell type 3 | |||
thymus epithelium | thymus epithelium | UBERON:0003483 | Hassall's corpuscle | Hassall's corpuscle | UBERON:0003987 | medullary thymic epithelial cell type 3 |
thymus epithelium | thymus epithelium | UBERON:0003483 | medullary thymic epithelial cell type 4; thymic tuft cells | |||
thymus epithelium | thymus epithelium | UBERON:0003483 | myo-medullary thymic epithelial cell | |||
thymus epithelium | thymus epithelium | UBERON:0003483 | neuro-medullary thymic epithelial cell |
UBERON:0003483 is thymus lymphoid tissue so is clearly used in error in the table where the ID for thymus epithelium should be used.
Please record here when fixed.
Can you make sure there are linked sub-issues for all the other issues in the breakdown/triage here? (We have tickets for at least some)
10-28-2021 removed 'Added new term' from ID column in thymus table v1.1 to eliminate error from error logs.
10-28-2021 replaced 7 instances of mismatched UBERON label with UBERON ID in ASCT+B thymus table v1.1 UBERON:0003483 is thymus lymphoid tissue
qualifier | author label | UBERON rdfs:label | UBERON ID |
---|---|---|---|
mismatch | thymus epithelium | thymus epithelium | UBERON:0003483 |
corrected | thymus epithelium | thymus epithelium | UBERON:0003846 |
For issue 3 listed here:
As we can see in this commit, all relations related to thymus lymphoid tissue
was removed from the table. In some of the cases, it was changed to thymus epithelium
.
Errors from error_ind_Thymus.log WARNING - Unrecognised cell content '' WARNING - Unrecognised cell content 'Added new term'
Author added unexpected content to ID column in table
Thymus 1.1. sheet
Tables Relationships in ASCT+B tables but not in Uberon Most are generic terms: Report generated on 29 September 2021