Evaluation of concordance of variant to phenotype associations in mouse and human.
In: Student Reports, Summer 2017, Jackson Laboratory
Dr. Cynthia Smith, Dr. Susan M. Bello and Debra M. Krupke
Epilepsy is a neurological disorder that is subdivided by syndrome. During the diagnosis a patient’s genome may be sequenced to determine treatment, identify the causative variant(s), and study these variants in a mouse model. The question asked in this study is what level of granularity (the phenotype, the gene variant, or the effect of the gene variant) will yield the most useful and accurate results for candidate gene analysis. Mouse models and human gene variants were compared in order to answer the question. Mouse Genome Informatics (MGI), Search Tool for the Retrieval of Interacting Genes/Proteins (STRING), and ClinVar were used to collect gene sets and gene variants for ion channel genes pathogenic in epilepsy. The pathogenic and likely pathogenic human variants were matched to mouse models by phenotype, sequence variant, type of mutation, and mutation location (exon). These data were analyzed by principal component analysis (PCA) and returned mutation type and mutation location as principal component 1 (PC1) and principal component 2 (PC2), respectively. Humanized mice are more informative than knockout of knockin mice. Studying disease moving forward researchers may want to focus on developing mouse models that mimic human gene variants by mutation location and mutation type to ensure models are as accurate as possible.
Alvarado, Cassandra, "Evaluation of concordance of variant to phenotype associations in mouse and human." (2017). Summer and Academic Year Student Reports. 2567.