Re: BioSamples type for review

Ah - didn’t know this. Great! 

Michael.

> On 13 May 2019, at 17:27, Chris Mungall <cjmungall@lbl.gov> wrote:
> 
> Prior to this I started on a document that compares different representations of samples across different schemas and domains, here it is if it's useful for the bioschemas effort:
> https://docs.google.com/document/d/1ExjXV5Y42SwH18cV91C9NbWkHooNjPpwL960lbT6eZ4/edit# <https://docs.google.com/document/d/1ExjXV5Y42SwH18cV91C9NbWkHooNjPpwL960lbT6eZ4/edit#>
>> Hi all,
>> 
>> would it make sense to align this between Phenopackets <-> bioschemas <-> schemablocks.org <http://schemablocks.org/> (well, I think it would, to a degree)?
>> 
>> - we are aware that geolocation data has to be provided, but not exactly about the extend (we think more about it in terms of “provenance”, not in a “where the hell is the sample stored”, which is outside of a data model’s scope, IMO)
>> - for material, with a focus on biomedical samples, we started to use new EFO classes w/ a target of “is this a sample representing the individual, or some abnormal part” (not yet in the schema)
>> - etc.
>> 
>> SchemaBlocks aligns (bi-directional) w/ Phenopackets, but tries to represent a larger set of scenarios. So any common parts (e.g. the format of geo data, if not the scoping) could be harmonised.
>> 
>> Michael.
>> 
>> 
>> 
>>  "provenance" : {
>>   "material" : {
>>    "type" : {
>>     "id" : "EFO:0009656",
>>     "label" : "neoplastic sample"
>>    }
>>   },
>>   "geo" : {
>>    "label" : "Antalya, Turkey",
>>    "precision" : "city",
>>    "city" : "Antalya",
>>    "country" : "Turkey",
>>    "latitude" : 36.91,
>>    "longitude" : 30.7
>>   }
>>  }