Bird phenotypes

[matentzn]

In todays call we decided tentatively to add bird phenotypes directly to uPheno.

I was also recommended to check out these resources and possibly reach out to @caripark who has some experience dealing with bird traits.

• http://avian.genomics.cn/en/
• https://b10k.genomics.cn/

• Animal QTL database - chicken traits

  @ImkeTammen will provide us with a list of bird phenotypes we will use as a seed for the project.

[cmungall]

remember there are lots of bird traits in oba already, we should coordinate and avoid ragged lattices

On Thu, Mar 24, 2022 at 11:48 AM Nico Matentzoglu @.***> wrote:

In todays call we decided tentatively to add bird phenotypes directly to uPheno.

I was also recommended to check out these resources and possibly reach out to @caripark https://github.com/caripark who has some experience dealing with bird traits.

• http://avian.genomics.cn/en/
• https://b10k.genomics.cn/
• Animal QTL database https://www.animalgenome.org/cgi-bin/QTLdb/index - chicken traits

@ImkeTammen https://github.com/ImkeTammen will provide us with a list of bird phenotypes we will use as a seed for the project.

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/upheno/issues/817, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAMMOI2QCKVDDPQZZIOZVLVBS2HDANCNFSM5RR753DQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[matentzn]

Sorry forgot to add that here to the ticket that these are of course coordinated, like everything else!

• OBA
• VT

But your comment about lattices - I don't get it in this context. Bird phenotypes is the exact same as any other phenotype ontology problem! The lattice wont be any more or less ragged!

[cmungall]

great! you are right in that if roughly the same requirements are driving oba as upheno the raggedness will be no worse

I don't know much much we have evaluated the raggedness, I expect we have a lot of:

• phenotype 1: very precise entity term, quite precise pato. term
• phenotype 2: slightly less precise entity term that ph1, slightly more precise pato term than in ph 1

where the common superclass of these is very much higher up

and I would expect this would be far worse across ontology (species-species, phenotype-trait) than within

On Thu, Mar 24, 2022 at 12:16 PM Nico Matentzoglu @.***> wrote:

Sorry forgot to add that here to the ticket that these are of course coordinated, like everything else!

• OBA
• VT

But your comment about lattices - I don't get it in this context. Bird phenotypes is the exact same as any other phenotype ontology problem! The lattice wont be any more or less ragged!

— Reply to this email directly, view it on GitHub https://github.com/obophenotype/upheno/issues/817#issuecomment-1077976624, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAMMONGIL5MSC6LANL4Q3DVBS5RLANCNFSM5RR753DQ . You are receiving this because you commented.Message ID: @.***>

[matentzn]

Our general approach has been to try and be more conservative on the PATO side - we only in the end use a handful of PATO terms (less than 100!) which cover most of the more generally useful phenotypes. In any case, thanks for mentioning this again, we will keep an eye on this!