Skip to content

Commit 817f5df

Browse files
authored
Fix minor errors in design pattern docs (NFDI4Chem#117)
* Fix typos and links * Update forgotten TBoxes and move logo to other file
1 parent ec7b411 commit 817f5df

6 files changed

+1289
-1436
lines changed

docs/VIBSO-logo.drawio

+586
Large diffs are not rendered by default.

docs/design_patterns.md

+21-31
Original file line numberDiff line numberDiff line change
@@ -72,7 +72,7 @@ To exemplify this pattern within the domain of VIBSO we can instantiate these cl
7272

7373
![OBI_asserted_assay_pattern_example.png](images/OBI_asserted_assay_pattern_example.png)
7474

75-
To better understand the defined logic of this pattern, it seems useful to look at its part more closely.
75+
To better understand the defined logic of this pattern, it seems useful to look at its parts more closely.
7676

7777
### Subjects of an Assay
7878
As we can see in this graph, the use of the term 'assay' is restricted to only those planned processes that examine a
@@ -144,77 +144,66 @@ relation between the 'setting datum' and the characteristic of the assay device
144144
quality'](https://terminology.nfdi4chem.de/ts/ontologies/vibso/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FPATO_0001241),
145145
which we can import from the Phenotype and Trait Ontology (PATO). It is the parent class of more specific physical
146146
characteristics, such as temperature, wavelength, mass or position, and it can be related to the device via the
147-
'quality of' relation from BFO. To be able to say that a 'setting datum' determines the value of such a device
147+
['quality of'](https://terminology.nfdi4chem.de/ts/ontologies/vibso/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FRO_0000080) relation from BFO. To be able to say that a 'setting datum' determines the value of such a device
148148
characteristic, we can use the RO relation 'causally influences'.
149149

150150
Since the actual quantitative value specified in a 'setting datum' can be encoded using different units of measurement,
151151
we also need a way to formally represent this. The way this is currently done in OBI is via the
152-
['value specification'](https://terminology.nfdi4chem.de/ts/ontologies/obi/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0001933)
153-
class, which is defined as an information "that specifies a value within a classification scheme or on a quantitative
154-
scale". We can link this class to a 'setting datum' via the 'has value specification' relation which is a
155-
specialization of the very general 'has part' relation. For the part of this value encoding that represents a given
156-
measurement unit, such as any SI unit, we can use one of the instances of the class 'unit' defined in the Unit
157-
Ontology (UO), and link it via the 'has measurement unit label' relation from IAO. The last missing pieces are the
158-
links between a value specification and the actual literal value respectively the device characteristic. Both are defined in
159-
OBI as ['has specified value'](https://terminology.nfdi4chem.de/ts/ontologies/obi/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0002135),
160-
respectively ['specifies value of'](https://terminology.nfdi4chem.de/ts/ontologies/obi/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0001927).
161-
152+
['value specification'](https://terminology.nfdi4chem.de/ts/ontologies/obi/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0001933) class, which is defined as an information "that specifies a value within a classification
153+
scheme or on a quantitative scale". We can link this class to a 'setting datum' via the ['has value specification'](https://terminology.nfdi4chem.de/ts/ontologies/vibso/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0001938)
154+
relation which is a specialization of the very general 'has part' relation. For the part of this value encoding that
155+
represents a given measurement unit, such as any SI unit, we can use one of the instances of the class '[unit](https://terminology.nfdi4chem.de/ts/ontologies/uo/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FUO_0000000)'
156+
defined in the Unit Ontology (UO), and link it via the ['has measurement unit label'](https://terminology.nfdi4chem.de/ts/ontologies/vibso/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FIAO_0000039) relation from IAO. The last
157+
missing pieces are the links between a value specification and the actual literal value respectively the device
158+
characteristic. Both are defined in OBI as ['has specified value'](https://terminology.nfdi4chem.de/ts/ontologies/obi/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0002135), respectively ['specifies value of'](https://terminology.nfdi4chem.de/ts/ontologies/obi/props?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FOBI_0001927).
162159

163160
Having described all the classes and relations we need to extend the OBI core assay pattern, we can now visualize
164161
this in from of the following graph:
165-
166162
![OBI_asserted_assay_pattern](images/OBO_setting_pattern.png)
167163

168164
When comparing this graph to the previous one, one can see that we have also added an 'is about' relation between
169165
'assay objective' and 'data item'. This relation was not added to the core pattern previously, as it is not formally
170166
asserted in OBI, but only given in the textual definition of the 'assay objective'.
171167

172168
To exemplify this extended pattern in the context of VIBSO, we can instantiate its classes as follows:
173-
174169
![OBO_setting_pattern_example.png](images/OBO_setting_pattern_example.png)
175170

176171
### Defining an Assay According to the Extended Assay Pattern in the Context of VIBSO
177172
As discussions with domain experts in the VIBSO development calls have made clear, the classification of
178-
the various types of Raman spectroscopy as currently done in CHMO has some problems (see https://github.com/NFDI4Chem/VibrationalSpectroscopyOntology/issues/103).
173+
the various types of Raman spectroscopy as currently done in CHMO has some problems (see [issue#103](https://github.com/NFDI4Chem/VibrationalSpectroscopyOntology/issues/103)).
179174
All of these types have in common that they somehow record the Raman scattering of the evaluated sample.
180175
However, their methodology differs in the ways in which the Raman scattering is produced, recorded or subsequently
181176
processed, by using either different devices, device settings, sample preparation steps or signal recording steps.
182177
We thus need the extended assay pattern described here, to be able to describe these different methods in more
183178
detail and thereby formally define the various types of Raman spectroscopy assays they specify.
184179

185-
186180
## The Assay Pattern in a Bigger Picture
187181
This assay pattern can of course combined with other planned process patterns, to be able to describe its embedding
188182
in a grander context. Without going into the details, the following TBox shows how this can be done with respect the
189183
competency questions:
184+
190185
* Who performed the assay?
191-
* what kind of data transformations where performed on the data output of the assay?
192-
* what kind of sampling process was done before the assay?
186+
* What kind of data transformations where performed on the data output of the assay?
187+
* What kind of sampling process was done before the assay?
193188
* What is the investigation the assay is a part of?
194189

195190
![OBI_asserted_assay_pattern](images/OBO_Investigation_Assay_Pattern.png)
196191

197-
198192
## Using Example Instances in VIBSO
199193
In the previous VIBSO development calls, it became apparent that the proper use of the design patterns described
200194
here should be illustrated with concrete example instances. This should not only make it easier for the domain
201195
experts with little knowledge on ontology development to follow. We also hope to be able to spot logical inconsistencies
202196
when minting new classes and asserting their class restrictions by using reasoners, such as ELK or HERMIT during the
203197
development.
204198

205-
We have thus added the ROBOT TSV template [scr/templates/vibso_examples.tsv](..%2Fsrc%2Ftemplates%2Fvibso_examples.tsv),
199+
We have thus added the ROBOT TSV template [vibso_examples.tsv](https://github.com/NFDI4Chem/VibrationalSpectroscopyOntology/blob/main/src/templates/vibso_examples.tsv),
206200
to be able to add example instances for each class minted in VIBSO. Similar to the other ROBOT templates we use,
207-
this one gets automatically converted into an OWL component ([vibso_examples.owl](..
208-
%2Fsrc%2Fontology%2Fcomponents%2Fvibso_examples.owl)) that is being imported into the main editor file ([vibso-edit.
209-
owl](..%2Fsrc%2Fontology%2Fvibso-edit.owl)).
201+
this one gets automatically converted into an OWL component ([vibso_examples.owl](https://github.com/NFDI4Chem/VibrationalSpectroscopyOntology/blob/main/src/ontology/components/vibso_examples.owl)) that is being imported into the
202+
main editor file ([vibso-edit.owl](https://github.com/NFDI4Chem/VibrationalSpectroscopyOntology/blob/main/src/ontology/vibso-edit.owl)).
210203

211-
-----
212-
213-
## External Sources Explaining and Reusing the Described Pattern
214-
215-
### OBI Quality & Quantity Pattern
204+
## OBI Quality & Quantity Pattern
216205
![measurement process pattern example](images/data_john_mass.png)
217-
Since we are reusing OBO ontologies and their patterns, we also try to reuse OBI's way of modeling
206+
Since we are reusing OBO ontologies and their patterns, we also try to reuse OBI's way of modeling
218207
[data and values](https://github.com/obi-ontology/obi/wiki/Data-and-Values). Please read their documentation for more
219208
background. Using this OBI pattern allows us to differentiate data values of qualities (aka attributes) of a material
220209
entity, such as the spectroscope or sample, into data values that represent settings and those that represent
@@ -224,11 +213,12 @@ Electronic Lab Notebooks, such a differentiation will most likely be very useful
224213
work in many OBO use cases, we need to see, if we have to adjust it for our needs. Other ontologies like QUDT or SIO
225214
use slightly different patterns to model qualities and their quantitative representations.
226215

227-
### Measurement Example KG of the Assay Pattern
216+
-----
217+
218+
## Measurement Example KG of the Assay Pattern
228219
Lars Vogt and Tobias Kuhn demonstrate the use of the OBO assay pattern within a grander research context as follows
229220
(cited from their preprint [DOI:10.13140/RG.2.2.13742.59203](http://doi.org/10.13140/RG.2.2.13742.59203), p.8):
230221
![measurement process pattern example](images/Fig4_10.13140_RG.2.2.13742.59203.png)
231-
232222
> Figure 3: A detailed machine-actionable representation of the metadata relating to a weight measurement datum
233223
> documented as an RDF ABox graph. The representation takes the form of an ABox semantic graph following the RDF
234224
> syntax. The graph documents a mass measurement process using a balance. It relates an instance of mass measurement
15.7 KB
Loading

docs/images/VIBSO_Raman_Tbox.png

-131 KB
Loading

0 commit comments

Comments
 (0)