Bad GPRs for Beta-Oxidation of Linoleoyl- and gamma-Linolenoyl-CoA Reactions

[Devlin-Moyer]

Current behavior:

I identified all of these problems while working on #761, and am putting them in a separate issue because they don't directly depend on the changes proposed in #761, and there's already a lot going on in #761. So:

• MAR03456, MAR03275, MAR03280, and MAR03284 should not be associated with ACADVL (ENSG00000072778) or ACADM (ENSG00000117054), because they involve unsaturated fatty acids, and ACADVL and ACADM can only act on saturated fatty acids (sources: this paper for ACADVL and the third paragraph of the discussion of this paper for MCAD)
• MAR03277, MAR03281, MAR03285, MAR03288, and MAR03290 should not be associated with EHHADH (ENSG00000113790), because these are mitochondrial reactions and EHHADH is peroxisomal (source: Uniprot)
  • update: MAR03288 was addressed by Remove EHHADH from the GPR of MAR03288 #790
  • update: MAR03277, MAR03281, MAR03285, and MAR03290 were addressed by fix: GPRs for mitochondrial 2-enoyl-CoA hydratase reactions #808
• MAR03278, MAR03282, MAR03286, and MAR03292 should not be associated with HADH (ENSG00000138796), because HADH can only act on substrates with 10 or fewer carbons in their main chains (source: Uniprot)

Proposed changes:

• Remove ENSG00000072778 and ENSG00000117054 from the GPRs of MAR03456, MAR03275, MAR03280, and MAR03284
• Remove ENSG00000113790 from the GPRs of MAR03277, MAR03281, MAR03285, MAR03288, and MAR03290
• Remove ENSG00000138796 from the GPRs of MAR03278, MAR03282, MAR03286, and MAR03292