Allow coloring structure at residues with uniprot variant mutations using DMS data

[Tom Goddard]

I'd like to be able to look at deep mutational scan scores for just residues for which there are clinically observed mutations for instance in the human mu opioid receptor. The observed mutations can be obtained from the current UniProt variants fetch. But in order to color using DMS scores the UniProt Variants mutation set needs to be merged with Willow's DMS data set in order to define an attribute that gives the DMS scores for just the clinical mutants.

A simple way to do this is at a command to merge mutation sets that have the same sequences.

I think a better way would be to allow mutation commands and guis to combine scores from different mutation sets provided they share a common sequence. If the score names are unique across all open mutation sets there would be no need for mutation set command options. Would also be nice to allow combined use of mutation sets with non-identical sequences such as Willow's DMS data with 10 sequences each differing by a single mutation from wild type. The restrictions on cross mutation set could be relaxed to only requires same residue numbering. All these better ways of combined use of mutation sets can be handle in other tickets.

[Tom Goddard]

Done.

I added a mutationscores merge command and also added an "observed" attribute to the mutation set returned from fetching Uniprot Variants. In order to show the DMS scores for only clinically observed mutations in a heatmap I allowed heatmaps to show computed scores. I added documentation for these new features.