HICF2: ALG13 S875N

HIFC2 entry with custom annotation ported to Michelanglo.

The patient presents Bilateral hippocampal sclerosis and a S875N mutation on the X-linked gene ALG13. This protein has two parts, both essential. One well-characterised and then other not and specific to animals.

Domain 1

ALG13 has two structured domains then a disordered one. The mutation is in the disordered region. The first domain is a glycosidase. It interacts with ALG14, which anchors it. Yeast have only this domain. The normal electrophoresis of transferrin is expected given that this domain is intact.
The SGC has been studing this as ALG13A. Nothing on the other isoform.

Domain 2 and C-terminus

However, from EcAC LoF mutations are lethal. The only cases are Q214* S149fs (×8) P162fs. Which are all before the second domain: OTU-like cysteine protease, which should be a deubiquitinase but lacks this activity or could not be detected. isoform 2 lacks this and C-terminal. So domain 2 is lethal without CTD. That is to say there are two activities in a single protein. The mutation is just before a polyproline helix.

The number of misses in ExAC is the same as random. So it seems unlikely to be a structured region. But it has to be complete. In ExAC there is a variant with Ser to Thr (not Asn). This is a very mild alteration.

Linear motif prediction of a loss of a potential SPOP binding site, which is linked to ubiquitination. STTS motif is conserved according to blast.

Looking at distribution of variants there is a lack of variants between 960-990 (LPPPPYSCDP SGSDLPQDTK VLQYYFNLGL). This could be disordered, but has a few potential binding sites. But what is it for?

Structure

The i-Tasser model is utter trash. The "best scoring" one gives a series of small globular domains, the one with the mutation, is from 865-919 (GAAANQAISTTSVSSQNAIQPLFVSPPTHGRPVIASPSYPCHSAIPHAGASLP).