#17 ECO: 0000028 motif similarity vs ECO:0000202

[alan_bridge]

Hi ECO

I wonder if you could take a look at the above mentioned term.

Definition=Used when an annotation is made based on the presence of a recognized domain or motif in a gene product's (usually protein) primary sequence.

It is not clear to me what the difference in usage and definition is between this evidence code and ECO:0000202, match to sequence model.

This term seems to try and separate "domain or motif" from other types of sequence model. However the difference between a family and a domain, or even a domain and a motif, is not always so clear cut, and may even vary between resources producting sequence models (like Pfam vs Prosite) or between resources using them (like UniProt or another sequence database). This could lead to inconsistent usage of this code ECO:0000202 and ECO:0000028.

Given the difficulties inherent in these definitions (family vs domain, domain vs motif) would you perhaps consider removing "motif similarity" and instead having one simple code for inferences based on matches to sequence models - ECO:0000202 - irrespective of whether the producer of the sequence model called it a domain or family?

Thanks and Best Regards, Alan

[Marcus]

Thank you for your comments, Alan.
We will consider your suggestion as we address the terms within similarity.
All the best,
Marcus