Modelling penetrance and severity

[ValWood]

Penetrance is displayed almost perfectly. Penetrance curation is a bit complicated by the fact that it can be a number, number range (percentage) or by a qualitative value "high" medium" or "low".

This is how it looks currently in the query output:

I think this works with the column headings.

For convenience, in PomBase, to make it easier to store the qualitative values in Chado, we converted these values to a "mini-ontology". However at PomBAse our users never see this. They only see the values. So, we should convert these into the text strings that they represent as follows: FYPO_EXT:0000001 = high FYPO_EXT:0000002 = medium FYPO_EXT:0000003 = low

This raises a question (for @kimrutherford ). Should these qualifiers be converted during loading into PomBemine. OR should we change export the text in the PHAF file instead of the IDs? Exporting the values might make more sense to all consumers of the PHAF file than the ontology term ID. This was really a workaround to make it easy to curate these annotation "qualifiers" rather than something that needed to be represented in an ontology with definitions etc?

[ValWood]

"Severity" is a qualifier very similar to penetrance except in this case there is only the text values "low" "medium" and high". It should be modelled in the same way as penetrance. Currrently it is modelled in a more complicated way:

and

For this we only need to store the qualifier value (Low, medium or high) as for penetrance above. Also the description of these terms does not make sense (it applies to both penetrance and severity, because we use the same "fake ontology" to store the values.

Apologies for not spotting this before. I only just figured out what was going on.

[kimrutherford]

Should these qualifiers be converted during loading into PomBemine. OR should we change export the text in the PHAF file instead of the IDs? Exporting the values might make more sense to all consumers of the PHAF file than the ontology term ID.

I think it would be OK to use "low", "high" etc. in the PHAF file. The Penetrance column already contains a bunch of non-term ID values (like "13" and "~82").

I've made a separate issue about it: pombase/pombase-chado#978

[ValWood]

OK that makes sense, it will make it simpler for @danielabutano

[kimrutherford]

I think it would be OK to use "low", "high" etc. in the PHAF file. The Penetrance column already contains a bunch of non-term ID values (like "13" and "~82").

That's done now, so both PHAF files have term names instead of term IDs for penetrance and severity:

• https://curation.pombase.org/dumps/latest_build/pombase-latest.eco.phaf.gz
• https://curation.pombase.org/dumps/latest_build/pombase-latest.phaf.gz

[danielabutano]

Ok, I have changed the model so both penetrance and severity will be String with value loaded directly from https://www.pombase.org/data/annotations/Phenotype_annotations `

<attribute name="severity" type="java.lang.String"/>
........

` I have already tried locally and sems working fine.

[ValWood]

Perfect!

[ValWood]

Perfect, both penetracne and severity are showing high, medium, low. Pentrance is also displaying percentage.