Propagate uniprot caution warnings to GO annotations (example from PARK7 gene)

[cmungall]

Example:

the PARK7 gene has been implicated in many things:
https://www.uniprot.org/uniprotkb/Q99497/entry

If we look at one publication, it suggests lactate biosynthesis via the methylglyoxal route:
https://amigo.geneontology.org/amigo/reference/PMID:22523093
(oddly, these annotations lack any annotation to a MF, but if the paper is to believed, the function is glyoxalase III activity, methylglyoxal + H2O = D-lactate)

However, if we look on the uniprot page we see there is a Caution for this publication:
https://www.uniprot.org/uniprotkb/Q99497/entry

The Caution can be retrieved by API to uniprot. I think it would be of service to GO users to propagate this where GO annotations are shown. This could also be shown in PAINT. Note this this paper has been used for IBA propagation despite its uncertain status: https://functionome.org/gene/UniProtKB:Q99497

The Caution notes could also be mined for NOTs.

This publication is also mentioned in the Caution, as it is the publication that called the original publication into question:
https://amigo.geneontology.org/amigo/reference/PMID:31653696

We do get a NOT for this, but I think the NOT is too specific.

[pgaudet]

• I dont see the NOT annotation in AmiGO ??

[cmungall]

The NOT is from the 2nd publication

https://amigo.geneontology.org/amigo/reference/PMID:31653696

The apparent deglycase activity of DJ-1 results from the conversion of free methylglyoxal present in fast equilibrium with hemithioacetals and hemiaminals

Snippet:

Several reports suggest that human DJ-1 is also capable of catalyzing the conversion of MGO and glyoxal, providing a potential explanation for the neuroprotective effects of DJ-1 (4, 5). It has been noted, however, that the glyoxalase activity of DJ-1 is much lower than that of other members of the DJ-1 superfamily (5, 6) and may be insufficient to allow for any meaningful physiological function in the human brain. The low glyoxalase activity of DJ-1 is supported by structural studies: DJ-1 lacks a histidine residue that is a part of the catalytic triad in Hsp31, making it a poor catalyst for efficient detoxification of MGO and glyoxal (7).

(reference 4 is the PMID with a caution association)

Here is the NOT:

[pgaudet]

@RLovering Can you have a look at these annotations?

[ValWood]

It would be amazing to harvest this negative data. We have a severe problem, one I deal with every day, that superseded science is rarely explicitly negated. Curators find this difficult to negotiate because, despite some early papers receiving no support, curators are often reluctant to remove earlier unsupported annotation, even in cases where there is no reproduction, or the results are clearly artefacts. I wish there was a way to encourage more publications with negative data (micropublications) to support NOTs.

There is a feature at Europe PMC that could be very cool. When (some) contrasting citations are identified

Unfortunately, this resource is only available by subscription (I have complained to Europe PMC about this multiple times as it is basically an advertisement).
An Open Source version of such a tool integrated into curation interfaces would be very useful.

[RLovering]

Hi

Rebecca added an alert to the GO annotations: PMID:25416785 show that the glyoxylase activity assigned to DJ-1/PARK7 in PMID:22523093 is actually the result of its deglycation activity. Therefore have edited my original (2014) annotation for this paper to remove the MF terms, and create more generic BP terms. There are no MF annotations for this paper.

The enzyme annotations associated with PARK7 from PMID:31653696 are SWIS annotations. What exactly did you want me to do?

Ruth