5.0.7.2 TO-DOs

[sammyjava]

• [x] GenotypingSample = Annotatable + description, name, organism, strain, dataSet
• [x] handle gzipped FASTAs and GFFs and TSVs (so I don't have to gunzip them from DS)
• • [x] lis-annotation
• • [x] lis-expression
• • [x] lis-fasta
• • [x] lis-genetic
• • [x] lis-map
• • [x] lis-markers
• • [x] lis-mstmap
• • [x] lis-synteny
• [x] add Gene<->ProteinDomain reverse reference as per @adf-ncgr
• [x] new GeneFamilyAssignment with Gene.geneFamilyAssignments, Protein.geneFamilyAssignments to support multiple gene family assignments per gene/protein

  <!-- stores a gene family assignment along with various measures -->
  <class name="GeneFamilyAssignment" is-interface="true">
  <attribute name="eValue" type="java.lang.Double"/>
  <attribute name="score" type="java.lang.Double"/>
  <attribute name="bestDomainScore" type="java.lang.Double"/>
  <reference name="geneFamily" referenced-type="GeneFamily"/>
  </class>

• [x] lis-annotation: associate protein with gene via transcript.ID=protein.ID rather than [gene.ID].n as per @adf-ncgr
• [x] lis-annotation: same deal with CDS (also loaded from a FASTA)

[sammyjava]

@adf-ncgr can I rely on the protein-named transcript being type=mRNA in the GFF? Or do I have to drill through other transcripty types?

[adf-ncgr]

I think you can rely on it (as much as anything else in gffland, anyway).

[sammyjava]

Collatoral Damage

• [x] SequenceAnnotateUtil
• [x] funnotate buttons - HACK: limit to gene family identifiers that start with "legfed" (whatever that is)
• [x] gene family enrichment widget - in webconfig so EACH MINE
• [x] funnotate list analysis widget - same HACK

[sammyjava]

In production as of https://github.com/legumeinfo/intermine/commit/891971b501ca450e11b2a9ca763ecf4b90534dc9