Closed cmungall closed 1 year ago
I agree these would be extremely useful for OBA.
Just to make sure I understand, OBA does not include classes for measurements, right, an these would map (somehow) to the relevant quality that is being measured?
@ramonawalls - while there is clearly a conceptual difference between "Total Basophil Count" measurement (an observation which has, for example, an observer), and "total basophil count" trait (an observable characteristic of an organism) we do not believe that having classes for both make any sense. It will take years to bridge "information" and "characteristic"-centric data models as they have proliferated widely, but it is our belief that only the "characteristic" itself should be represented as a named class (the "measurement" aspect of a characteristic is purely a "shadow" of the actual characteristic), and that the occurrence of either one (measurement or trait) should be, for most biomedical data integration use cases, "conflated" (i.e. you should pretend they are the same). So in this case here (as with our EFO alignment) we will create all of these as trait terms, but pretend, that they relate to the same thing (i.e yes, map across).
Thanks, @matentzn . I agree completely that there should not be parallel hierarchies of traits and measurement types. Saying that there is "a conceptual difference" between the two types understates the importance of the difference, at least for the kind of work I have done, but I support the conflation approach for knowledge graph work. I just wanted to be sure that in OBA, qualities are logically defined as qualities, not processes or information, so that if we want to use more robust reasoning, we can.
C-Path should be able to commit some resources to the mapping, if you can let us know what needs to be done.
I just wanted to be sure that in OBA, qualities are logically defined as qualities, not processes or information, so that if we want to use more robust reasoning, we can.
100%! This is what they are.
@rays22 hold of on this for a moment.
@ramonawalls basically what we need is simply a two column table for each of these NCIT terms, the first column being the appropriate PATO quality and the second the respective entity, i.e. Hematocrit etc. This is not always trivial, but I think its not too bad.
The ratio ones require a new pattern I think. @rays22 can you propose a simple logical pattern for capturing the ratio traits? It should be aligned with uPheno.
Here is a mapping table of 30 NCIT measurement terms:
NCIT Identifier | NCIT label | predicate | OBA ID | OBA label |
---|---|---|---|---|
NCIT:C64488 | Calcium Measurement | skos:exactMatch | OBA:VT0010499 | calcium amount |
NCIT:C61033 | Serum Total Protein Measurement | skos:exactMatch | OBA:VT0005567 | blood total protein amount |
NCIT:C61019 | Blood Urea Nitrogen Measurement | skos:exactMatch | OBA:VT0005265 | blood urea nitrogen amount |
NCIT:C51951 | Platelet Count | skos:exactMatch | OBA:VT0003179 | platelet quantity |
NCIT:C64550 | Eosinophil Count | skos:exactMatch | OBA:VT0002602 | eosinophil quantity |
NCIT:C61031 | Serum Total Bilirubin Measurement | skos:exactMatch | OBA:VT0001569 | blood bilirubin amount |
NCIT:C51949 | Lymphocyte Count | skos:exactMatch | OBA:VT0000717 | lymphocyte quantity |
NCIT:C51950 | Neutrophil Count | skos:exactMatch | OBA:VT0000222 | neutrophil quantity |
NCIT:C51948 | Leukocyte Count | skos:exactMatch | OBA:VT0000217 | leukocyte quantity |
NCIT:C61023 | Serum Creatinine Measurement | skos:exactMatch | OBA:2050096 | serum creatinine amount |
NCIT:C61015 | Serum Albumin Measurement | skos:exactMatch | OBA:2050068 | serum albumin amount |
NCIT:C61017 | Serum Alanine Aminotransferase Measurement | skos:exactMatch | OBA:2050062 | serum alanine aminotransferase amount |
NCIT:C64796 | Hematocrit Measurement | skos:exactMatch | OBA:2045381 | hematocrit |
NCIT:C61037 | Serum Triglyceride Measurement | skos:exactMatch | OBA:2045035 | level of triglyceride in blood serum |
NCIT:C61016 | Serum Alkaline Phosphatase Measurement | skos:exactMatch | OBA:2040575 | level of alkaline phosphatase, tissue-nonspecific isozyme in blood serum |
NCIT:C64495 | Chloride Measurement | skos:exactMatch | OBA:1000959 | chloride amount |
NCIT:C64545 | Carbon Dioxide Measurement | skos:exactMatch | OBA:1000957 | carbon dioxide amount |
NCIT:C64470 | Total Basophil Count | skos:broadMatch | OBA:VT0002607 | basophil quantity |
NCIT:C61018 | Serum Aspartate Aminotransferase Measurement | skos:narrowMatch | OBA:2041767 | level of aspartate aminotransferase, mitochondrial in blood serum |
NCIT:C61018 | Serum Aspartate Aminotransferase Measurement | skos:narrowMatch | OBA:2045394 | level of aspartate aminotransferase, cytoplasmic type in blood serum |
NCIT:C61029 | Serum Sodium Measurement | skos:exactMatch | OBA:2045395 | serum sodium level |
NCIT:C61027 | Serum Glucose Measurement | skos:exactMatch | OBA:2045396 | serum glucose level |
NCIT:C61041 | Serum HDL Cholesterol Measurement | skos:exactMatch | OBA:2045397 | serum high-density lipoprotein cholesterol level |
NCIT:C61042 | Serum LDL Cholesterol Measurement | skos:exactMatch | OBA:2045398 | serum low-density lipoprotein cholesterol level |
NCIT:C64433 | Alanine Aminotransferase Measurement | skos:exactMatch | OBA:2045399 | alanine aminotransferase level |
NCIT:C64467 | Aspartate Aminotransferase Measurement | skos:narrowMatch | OBA:2045400 | level of aspartate aminotransferase, cytoplasmic type |
NCIT:C64467 | Aspartate Aminotransferase Measurement | skos:narrowMatch | OBA:2045401 | level of aspartate aminotransferase, mitochondrial type |
NCIT:C64547 | Creatinine Measurement | skos:exactMatch | OBA:2045402 | creatinine level |
NCIT:C64816 | Urobilinogen Measurement | skos:exactMatch | OBA:2045403 | urobilinogen level |
NCIT:C64810 | Nitrite Measurement | skos:exactMatch | OBA:2045404 | nitrite level |
NCIT:C64471 | Basophil to Leukocyte Ratio | skos:broadMatch | OBA:2045390 | basophil ratio |
NCIT:C64604 | Eosinophil to Leukocyte Ratio | skos:broadMatch | OBA:2045391 | eosinophil ratio |
Examples
NCIT:C51948 ! Leukocyte Count NCIT:C51949 ! Lymphocyte Count NCIT:C51950 ! Neutrophil Count NCIT:C51951 ! Platelet Count NCIT:C61015 ! Serum Albumin Measurement NCIT:C61016 ! Serum Alkaline Phosphatase Measurement NCIT:C61017 ! Serum Alanine Aminotransferase Measurement NCIT:C61018 ! Serum Aspartate Aminotransferase Measurement NCIT:C61019 ! Blood Urea Nitrogen Measurement NCIT:C61023 ! Serum Creatinine Measurement NCIT:C61027 ! Serum Glucose Measurement NCIT:C61029 ! Serum Sodium Measurement NCIT:C61031 ! Serum Total Bilirubin Measurement NCIT:C61033 ! Serum Total Protein Measurement NCIT:C61037 ! Serum Triglyceride Measurement NCIT:C61041 ! Serum HDL Cholesterol Measurement NCIT:C61042 ! Serum LDL Cholesterol Measurement NCIT:C64433 ! Alanine Aminotransferase Measurement NCIT:C64467 ! Aspartate Aminotransferase Measurement NCIT:C64470 ! Total Basophil Count NCIT:C64471 ! Basophil to Leukocyte Ratio NCIT:C64488 ! Calcium Measurement NCIT:C64495 ! Chloride Measurement NCIT:C64545 ! Carbon Dioxide Measurement NCIT:C64547 ! Creatinine Measurement NCIT:C64550 ! Eosinophil Count NCIT:C64604 ! Eosinophil to Leukocyte Ratio NCIT:C64796 ! Hematocrit Measurement NCIT:C64810 ! Nitrite Measurement NCIT:C64816 ! Urobilinogen Measurement
Note: as stated a few other places I don't think we should separate characteristics from measurements, it's the same concept
I believe this would be of value to cpath cc @nicolevasilevsky