ValWood
commented
2 years ago (last week) we chatted about how 'reasoning ' can be used to instantiate annotations for "mitotic spindle pole" if annotations exist for "mitotic spindle pole body" and "during anaphase, anaphase A, anaphase B"
There is a precedent for using the ontology relations to generate missing annotations. I documented the existing, and types proposed here:
https://docs.google.com/document/d/11oyTCn4xxGyfr_Q9geD4ZFGmlqUxg3wtK3GX97gYxSQ/edit
For now, I can easily create the missing annotations for the genes which have "mitotic spindle pole body" during ""mitotic anaphase" and add them to the legacy GO file (note this file is misnamed, it includes all of the manual annotations that cannot be made in Canto (ISS,IC,TAS,NAS) in additional to legacy IMP,IGI,IPI
All I need to do is grep GO:0044732 >gene_association_file> |grep -v HDA | grep GO:0000091 > mitotic_spindle_pole grep GO:0044732 >gene_association_file> |grep -v HDA | grep GO:0000090 >>mitotic_spindle_pole grep GO:0044732 >gene_association_file> |grep -v HDA | grep GO:0000092 >>mitotic_spindle_pole perl -pi -e 's/GO:0044732/GO:0097431/g' mitotic_spindle_pole Add the annotations to the legacy gaf file pombe-embl/supporting_files/legacy_go_annotations_from_contigs.txt
@manulera Could you have a look through the list of annotations to mitotic spindle pole body (ignoring the HTP), and see if any could have a "during anaphase" label , then I will make the annotations.
Based on the current data, we only et only 20 annotations for 17 genes
cc @ValWood
One of my favourite papers, very mechanistic about merotelic attachments (when a given kinetochore is attached to both spindle poles so when anaphase starts it does not segregate to either pole and lags).
They provide mechanistic evidence for the model above, basically that if there are merotelic attachments during anaphase B spindle elongation:
Below is what we have in FYPO so far, I think it looks quite good for the single mutant phenotypes. We would have to add something extra for the special collapse of spindle in the double mutant (novel in this paper). I wonder if in GO some causal links can be made between these processes, since this paper provides very good mechanistic evidence. The other thing with GO is that these mechanisms might not be either conserved or be as relevant. It's quite accepted that stretching kinetochore will help resolve merotelic attachments, but the spindle collapse observed in pombe might not happen if you delete ase1 orthologs in other organisms.
Below this graph, there is also a GO graph of spindle elongation, but it does not seem any of these connections between the processes exists.