#204 import extra info for ortholog table and write out at update time

[Valerie Wood]

New ortholog (version 19 is made)
https://sourceforge.net/p/pombase/curation-tasks/208/

All of the genes which are tandem fusions, or map to a tandem fusion are annotated thus:

/controlled_curation="term=orthologous to S. cerevisiae YDR267C; qualifier=SPAC806.02c,C_term; date=19700101"
/controlled_curation="term=orthologous to S. cerevisiae YIL003W; qualifier=SPAC806.02c,N_term; date=19700101"

OR
tandem fusion in S. cerevisia
his7
/controlled_curation="term=orthologous to S. cerevisiae YCL030C; qualifier=YCL030C(N-term); date=19700101"
his2
/controlled_curation="term=orthologous to S. cerevisiae YCL030C; qualifier=YCL030C(C-term); date=19700101"

So next you need to
1. Make sure "qualifier= is parsed and stored (this might happen already)
2. Make sure db_xref is parsed and stored
Let us know if this is happening as we can open a jira ticket to display these

1. Write out an equivalent version (V.19) from Chado
   (I will compare this to my existing V 19 to amke sure nothing has gon awry, because the table is currently maintained independently from the data in Artemis).
   Once I have checked it, this will become the official release.

[Valerie Wood]

There is now also one example for human:
ade5 /controlled_curation="term=human GART ortholog; qualifier=HGMP:GART,C-term; date=20130815"
ade1 /controlled_curation="term=human GART ortholog; qualifier=HGMP:GART,N-term; date=20130815"

[Valerie Wood]

Also when these table sare written out, they should exclude "dubious genes"
(currently included in both the human table, and in my manual pombe table)

[Valerie Wood]

When a cluster occurs , make sure it has the same gene combination each time it occurs like this

SPAC1002.17c | UCK1 UCK2 UCKL1
SPAC1399.04c | UCK1 UCK2 UCKL1

not like this:

SPAC1002.17c | UCK1 UCK2 UCKL1
SPAC1399.04c | UCK1 UCK2

[Valerie Wood]

pasting this ole e-mail exchange in here so I don't forget... but it may well be on the tracker somewhere. don't worry about this one...I will go through all of the ortholog table related tickets soon and figure out what still needs to be done and write a sensible ticket.

Val

==
Hi Kim,
I wonder if when you write out the human ortholog table, the Human Id lists could be alphabetised.
I have a list of cut and grep command I can use on the cerevisiae ortholog table to generate a table
of 1:1,
1:many
many:1
many:many.

It doesn't work on the human table becasue it depends on sorting similar, so these currently sort individually even though they have the same orthologs:

1 ATAD2B|ATAD2
1 ATAD2|ATAD2B

1 RPL10L|RPL10
1 RPL10|RPL10L

1 SLC17A1|SLC17A6|SLC17A2|SLC17A3|SLC17A8|SLC17A5|SLC17A7|SLC17A4
1 SLC17A1|SLC17A6|SLC17A3|SLC17A2|SLC17A4|SLC17A7|SLC17A5|SLC17A8
1 SLC17A1|SLC17A6|SLC17A3|SLC17A2|SLC17A5|SLC17A8|SLC17A4|SLC17A7
1 SLC17A3|SLC17A2|SLC17A6|SLC17A1|SLC17A5|SLC17A8|SLC17A4|SLC17A7
1 SLC17A3|SLC17A2|SLC17A6|SLC17A1|SLC17A8|SLC17A5|SLC17A4|SLC17A7
1 SLC17A4|SLC17A7|SLC17A5|SLC17A8|SLC17A3|SLC17A2|SLC17A6|SLC17A1
1 SLC17A4|SLC17A7|SLC17A5|SLC17A8|SLC17A6|SLC17A1|SLC17A3|SLC17A2
1 SLC17A5|SLC17A8|SLC17A7|SLC17A4|SLC17A1|SLC17A6|SLC17A2|SLC17A3
1 SLC17A8|SLC17A5|SLC17A4|SLC17A7|SLC17A3|SLC17A2|SLC17A1|SLC17A6

[Valerie Wood]

note: in these instances

/gene="SPAC806.02c"
/controlled_curation="term=human NUPBP2 ortholog; qualifier=SPAC806.02c,N-term; date=20140831"
/controlled_curation="term=human CIAO1 ortholog; qualifier=SPAC806.02c,C-term; date=20140831"

only the second one is getting thorugh to the gene pages right now. However its only a very few like this for human so it can be dealt with with this ticket after multigene phenotypes.

v