Bad GPRs for Mitochondrial delta-3-delta-2-enoyl-CoA Isomerization Reactions

[Devlin-Moyer]

Current behavior:

These reactions all represent delta-3- delta-2-enoyl-CoA isomerization in the mitochondria and have GPRs of "ECI1 or ECI2 or EHHADH":

ID	Reaction
MAR03288	cis,cis-3,6-dodecadienoyl-CoA --> trans,cis-lauro-2,6-dienoyl-CoA
MAR03374	trans-2-cis,cis-5,8-tetradecatrienoyl-CoA <=> (3E,5Z,8Z)-Tetradecatrienoyl Coenzyme A
MAR03378	Trans-2-Cis-8-Tetradecadienoyl Coenzyme A <=> Trans-3-Cis-8-Tetradecadienoyl Coenzyme A

Expected feature/value/output:

While all three of these enzymes can catalyze this type of reaction, EHHADH (ENSG00000113790) is only localized to peroxisomes (ECI1 is localized to mitochondria, and ECI2 can be localized to either, depending on how it's spliced). EHHADH should be removed from all of these GPRs.

EHHADH is also associated with a lot of other mitochondrial reactions, but I'm trying to address them in small groups to avoid large complicated issues.

Proposed changes:

• Remove ENSG00000113790 from the GPR of MAR03288
• Remove ENSG00000113790 from the GPR of MAR03374 addressed in fix: remove genes only expressed in peroxisome from beta-oxidation reactions in mitochondria #770
• Remove ENSG00000113790 from the GPR of MAR03378 addressed in fix: remove genes only expressed in peroxisome from beta-oxidation reactions in mitochondria #770

[haowang-bioinfo]

nice findings and thanks

[Devlin-Moyer]

fixed in #790