PAINT pipeline probem

[ValWood]

Something is very wrong here.

Pombe Lto1 https://www.pombase.org/gene/SPCC191.08 is http://www.pantherdb.org/panther/family.do?clsAccession=PTHR28532

protein maturation by [4Fe-4S] cluster transfer ribosomal large subunit biogenesis

has human ortholog LTO1 http://www.uniprot.org/uniprot/Q8WV07 this is getting IBA annotations to double-strand break repair via homologous recombination Source: GO_Central intra-S DNA damage checkpoint Source: GO_Central meiotic DNA integrity checkpoint Source: GO_Central mitotic DNA replication checkpoint Source: GO_Central nucleotide-excision repair Source: GO_Central (all incorrect)

from a totally different family PANTHER:PTN000307391 which includes annotations transferred from pome SPAC20G4.04c|UniProtKB=P78955 hus1 (which is a checkpoint protein)

hus1 and lto1 are unrelated?

@pgaudet @selewis @huaiyumi

[ValWood]

@pgaudet maybe you could check if this one is reolved. The pipline might be ironed out by now.

[ValWood]

human LTO1 still seems to be picking up mappings from the wrong tree. I can't figure it out... http://www.pantree.org/node/annotationNode.jsp?id=PTN000307391

[ValWood]

This is pom be LTO1 http://amigo.geneontology.org/amigo/gene_product/PomBase:SPCC191.08 it does not have the PANTHER family.

human LTO1 is missign from AMigo?

This is mouse LTO1 http://amigo.geneontology.org/amigo/gene_product/MGI:MGI:1919534

It is in a family which maps it to pombe https://www.pombase.org/spombe/result/SPAC20G4.04c but this is not related to LTO1

[ValWood]

See the LTO1 orthologs https://www.flyrnai.org/cgi-bin/DRSC_orthologs.pl?input_species=4896&output_species=0&additional_filter=None&search_fields=***&search_datasets=***&gene_list=SPCC191.08

Note, that DIOPT reports ORAOV1 but this is not called LTO1 in human rat and mouse

[ValWood]

This is the human entry in AmiGO. It still has the old name http://amigo.geneontology.org/amigo/gene_product/UniProtKB:Q8WV07 but it does have the incorrect mappings that come from a seemingly unrelated family

[pgaudet]

@ValWood

pombe lto1 has a single IBA

Are you describing multiple problems ?

Thanks, Pascale

[pgaudet]

I think I get it. Q8WV07 is in fact in PTHR12900 (maybe the families have changed?) as an parlog Hus1, so perhaps we cannot be sure of the function. I'll remove these annotations for Q8WV07.

Should I also remove the annotations to HusB ?

[pgaudet]

I removed all annotations to the ORAOV1 branch, if that works for you please close.

Pascale

[ValWood]

I think there is a problem with the family though.

pombe LTO1 https://www.pombase.org/gene/SPCC191.08 ORAOV1 ortholog

is not related (at all) to pombe hus1 https://www.pombase.org/gene/SPAC20G4.04c HIUS1 ortholog

Lto1 is a http://pfam.xfam.org/family/PF09811#tabview=tab2 Hus1 is a ATPase checkpoint protein http://pfam.xfam.org/family/PF04005#tabview=tab2 They are not even in the same clan.

Maybe there is another domain below threshold in one of them. But to me they seem to be totally unrelated.

I think this one panther need to look at.

[pgaudet]

I sent an email to @huaiyumi , he'll look into this for the next Panther release.