Gene Validity Classification Summary

Gene/Disease Pair:

MAGI2 : infantile epilepsy syndrome

HGNC:18957 | MONDO_0020071
Mode of Inheritance: X-linked inheritance (HP:0001417)
Expert Panel: Epilepsy EP
SOP: Gene Clinical Validity Standard Operating Procedures (SOP), Version 5

Genetic Evidence
Case-Level Data
Evidence Type Case Information Type Guidelines Points PMIDs/Notes
Default Range Max Total Counted
Variant Evidence
Autosomal Dominant or X-linked Disorder Variant is de novo 2 0-3 12
Proband with predicted or proven null variant 1.5 0-2 10 0 0
Bierzynska A et al. 2017 May (PMID:27932480);
Proband with other variant type with some evidence of gene impact 0.5 0-1.5 7
Autosomal Recessive Disease Two variants in trans and at least one de novo or a predicted/proven null variant 2 0-3 12
Two variants (not predicted/proven null) with some evidence of gene impact in trans 1 0-1.5
Segregation Evidence Evidence of segregation in one or more families   Sequencing Method 0-3 3
Total LOD Score Canditate Gene Sequencing Exome/Genome or all genes sequenced in linkage region  
2-2.99 0.5 1
3-4.99 1 2
≥5 1.5 3
Case-Control Data
Case-Control Study Type Case-Control Quality Criteria Guidelines Points PMIDs/Notes
Points/Study Max Total Counted
Single Variant Analysis 1. Variant Detection Methodology
2. Power
3. Bias and confounding
4. Statistical Significance
0-6 12
Aggregate Variant Analysis 0-6
Total Genetic Evidence Points (Maximum 12) 0
Experimental Evidence
Evidence Category Evidence Type Guidelines Points PMIDs/Notes
Default Range Max Total Counted
Function Biochemical Function 0.5 0 - 2 2
Protein Interaction 0.5 0 - 2 0
Deng F et al. 2006 Jul 26 (PMID:16870733); Bauß K et al. 2014 Aug 1 (PMID:24608321);
Expression 0.5 0 - 2 0.5
Balbas MD et al. 2014 Oct 14 (PMID:25271328); Deng F et al. 2006 Jul 26 (PMID:16870733);
Functional Alteration Patient cells 1 0 - 2 2
Non-patient cells 0.5 0 - 1
Models Non-human model organism 2 0 - 4 4 0 0
Ihara K et al. 2014 Oct (PMID:25108225); Komoike Y et al. 2010 Apr (PMID:20146355);
Cell culture model 1 0 - 2
Rescue Rescue in human 2 0 - 4
Rescue in non-human model organism 2 0 - 4
Rescue in cell culture model 1 0 - 2
Rescue in patient cells 1 0 - 2
Total Experimental Evidence Points (Maximum 6) 0.5



Assertion criteria Genetic Evidence (0-12 points) Experimental Evidence
(0-6 points)
Total Points
Replication Over Time (Y/N)
Description Case-level, family segregation, or case-control data that support the gene-disease association Gene-level experimental evidence that support the gene-disease association Sum of Genetic & Experimental
> 2 pubs w/ convincing evidence over time (>3 yrs)
Assigned Points 0 0.5 0.5 NO
STRONG 12-18
DEFINITIVE 12-18 AND replication over time
Valid contradictory evidence (Y/N)*
There are at least 4 individuals that have been identified who have a deletion on the same locus with infantile spasms but are not missing MAGI2
MAGI2 was evaluated for evidence supporting and refuting its relationship with infantile epilepsy (MONDO:0020071). Of note, variation in MAGI2 has also been described in association with Nephrotic Syndrome, an AR disorder; evidence for this gene-disease relationship is not considered here. MAGI2 was initially thought to be involved in infantile epilepsy after reports of large deletions including MAGI2 (and other genes) in individuals with infantile epilepsy (Marshall 2008 18565486, Peterson 26030165). MAGI2 was thought to be the causative gene because it has been found to interact with stargazin, a protein that has been associated with epilepsy in mice (Deng 2006 16870733), and it was the region with the most overlap amongst patients with documented infantile spasms (Marshall 2008 18565486). However, no single-gene deletions or single nucleotide variants have been reported in this gene in individuals with infantile epilepsy; additionally, the Marshall 2008 and Rothlisberger 2010 studies each contained one individual with a deletion at the 7q11.23-7q2.11 locus that did not encompass MAGI2. There has also been an individual with a deletion at this locus that spanned part of MAGI2 who did not have a a diagnosis of epilepsy though it had not been ruled out (Marshall 2008 18565486). Additionally, there have been 4 other individuals with deletions at this locus that do not include MAGI2 who presented with infantile spasms suggesting that there is a different candidate gene for this disorder (Paciorkowski 21694734, Morimoto 2003 14636357, Komoike 20146355). Furthermore, the MAGI2 null mouse displayed renal abnormalities consistent with patients with intragenic null frameshift variants who had no reports described of seizures indicating that the absence of MAGI2 may cause nephrotic syndrome instead of infantile epilepsy (Ihara 2014 25108225, Bierzynska 2017 27932480). In conclusion, given that there have been no variants impacting only MAGI2 in patients with infantile spasms, patients with MAGI2 variants have expressed a different phenotype, and patients with deletions at this locus that do not include MAGI2 have reported seizures, this gene-disease association is Disputed.