search

obi-ontology / obi

The Ontology for Biomedical Investigations
http://obi-ontology.org
Creative Commons Attribution 4.0 International
75 stars 26 forks source link

NTR: RNA integrity number #824

Open cstoeckert opened 7 years ago

cstoeckert commented 7 years ago

The Ontology for Biobanking (OBIB) has requests from the NCI Biorepositories and Biospecimen Research Branch (BBRB) that include RIN (RNA integrity number) values associated with specimens. I propose adding this to OBI as: Label: ‘RNA integrity number’ Alternative label: RIN Definition: A data item that is calculated using an algorithm to assess RNA quality based on features from an electrophoretic trace. Values range from 10 (most intact) to 1 (most degraded). Definition source: https://www.agilent.com/cs/library/applications/5989-1165EN.pdf Ontology term requestor: NCI BBRB.

bpeters42 commented 7 years ago

As we are dealing with it, should we take these as examples for how to deal with value specifications? These seem to be great examples. You can both use a RIN number in a plan (e.g. discard samples with RIN < 7), and a measured data item of a RIN number for a specific sample. The values come from a restricted range (1-10). And it isn't a scalar with unit in the classical sense.

I guess with that I am saying: Should we define the RIN number value specification instead?

On Tue, May 30, 2017 at 9:25 AM, cstoeckert notifications@github.com wrote:

The Ontology for Biobanking (OBIB) has requests from the NCI Biorepositories and Biospecimen Research Branch (BBRB) that include RIN (RNA integrity number) values associated with specimens. I propose adding this to OBI as: Label: ‘RNA integrity number’ Alternative label: RIN Definition: A data item that is calculated using an algorithm to assess RNA quality based on features from an electrophoretic trace. Values range from 10 (most intact) to 1 (most degraded). Definition source: https://www.agilent.com/cs/library/applications/5989- 1165EN.pdf Ontology term requestor: NCI BBRB.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/824, or mute the thread https://github.com/notifications/unsubscribe-auth/ANN9IgT_sg1UIq4dlqg-Xjj91nRmSxg6ks5r_EL5gaJpZM4NqiRi .

-- Bjoern Peters Associate Professor La Jolla Institute for Allergy and Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

cstoeckert commented 7 years ago

A RIN is produced by a (validated) method (algorithm) that produces fractional values (are not categorical) so fits the definition of a data item. RIN values, however, can be specified as real numbers between 1 and 10. Therefore I think one can define a value specification for RIN values but I don't think it make sense to do so without first defining a RIN.

bpeters42 commented 7 years ago

I guess I am trying to say: Should we define: "RIN number" as a kind of 'value specification' rather than as a kind of data item? This would be part of the broader redesign, picking apart what the form of the value specification is (here, a real number, 1-10, and intended as an indicator of RNA quality), and how it is used (as a data item vs. e.g. as a threshold of minimal RNA-quality in a plan specification).

I am not saying that all of this is well worked out right now, and as an intermediate I am fine with just adding it as a data item if you need it immediately.

On Tue, May 30, 2017 at 3:07 PM, cstoeckert notifications@github.com wrote:

A RIN is produced by a (validated) method (algorithm) that produces fractional values (are not categorical) so fits the definition of a data item. RIN values, however, can be specified as real numbers between 1 and

  1. Therefore I think one can define a value specification for RIN values but I don't think it make sense to do so without first defining a RIN.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/824#issuecomment-305022692, or mute the thread https://github.com/notifications/unsubscribe-auth/ANN9Ik1GDGBJNAhmHxUjlP187Uu68Qsoks5r_JMNgaJpZM4NqiRi .

-- Bjoern Peters Associate Professor La Jolla Institute for Allergy and Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

cstoeckert commented 7 years ago

Thanks, unless I hear other issues, I will work on adding RIN as data item. To flesh out my thinking further on this, a data item should be defined by what generated it. That is what I need for a RNA Integrity Number (generated by the RIN algorithm). In my use case, I need to be able to say specifically that there was a RNA Integrity Number generated not how it is going to be used which is certainly important but different. A value specification should define what are allowable values for a given purpose and not necessarily what generated them.

cstoeckert commented 7 years ago

Based on the discussions from the June 12, 2017 OBI call, now propose the following: RNA Integrity Number calculation: a data transformation using the RIN algorithm to generate a quality measure of RNA based on features from an electrophoretic trace. Alternative label: RIN calculation Definition source: https://www.agilent.com/cs/library/applications/5989-1165EN.pdf Ontology term requestor: NCI BBRB. has specified output some (measurement datum and has some RNA Integrity Number value specification).

RNA Integrity Number value specification: a value specification that specifies the value of the RNA Integrity Number as a real value between 1 (most degraded) and 10 (most intact). Alternative label: RIN value specification.

DanBerrios commented 7 years ago

It's not always done by electrophoresis. Commonly done by spectrophotometry too.

Dan

bpeters42 commented 7 years ago

Perfect! It does look like we are completely on the same page.

On Tue, Jun 13, 2017 at 1:38 PM, cstoeckert notifications@github.com wrote:

Based on the discussions from the June 12, 2017 OBI call, now propose the following: RNA Integrity Number calculation: a data transformation using the RIN algorithm to generate a quality measure of RNA based on features from an electrophoretic trace. Alternative label: RIN calculation Definition source: https://www.agilent.com/cs/library/applications/5989- 1165EN.pdf Ontology term requestor: NCI BBRB. has specified output some (measurement datum and has some RNA Integrity Number value specification).

RNA Integrity Number value specification: a value specification that specifies the value of the RNA Integrity Number as a real value between 1 (most degraded) and 10 (most intact). Alternative label: RIN value specification.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/824#issuecomment-308241301, or mute the thread https://github.com/notifications/unsubscribe-auth/ANN9Ih6k9htKvfDU4Xsq4T5xbQF7W8oGks5sDvM6gaJpZM4NqiRi .

-- Bjoern Peters Associate Professor La Jolla Institute for Allergy and Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

cstoeckert commented 7 years ago

Added and committed. Still need to address Dan's point and add annotation property that this from NCI BBRB.