Support for search based on Domain Architecture

[daviortega]

Why?

MiST2 used to have a search box to find proteins with a certain domain. It was a very useful tool and I think we should bring it to MiST3.

How (ideally)?

Approach 1

My first instinct is that we should make a POST route where the advanced users can post RechArch definition and a taxonomic context.

The route could filter the aseqs present in the taxonomy context and then compute the RegArch on them.

It might not be ideal to compute RegArch to ALL Aseqs every time. Thus, we could include a step in the pipeline and a new DB table to index the protein features and associate a list of aseq with it.

That would speed up the process of making a POST call with RegArch, since we could pre-process the request and only apply the RegArch to the subset of aseqs with domains that are part of the RegArch request.

UI

For the user using the GUI, we can keep it simple and only allow requests for the presence of a single domain within a genome or not. The GUI generates a simple RegArch and sends it on the POST route. The answer is rendered as a list of genes like the GUI does already.

---

These are half baked thoughts, but we should definitely discuss how to make this option available in MiST3.

It would be nice to get the opinion of @lukeulrich, @adebali and @ToshkaDev about this.