Construction of Computational Pipeline to Analyze Differences in Gene Expression in Inbred Mouse Strains


Maria Flores

Document Type


Publication Date


JAX Location

In: Student Reports, Summer 2019, The Jackson Laboratory


Inbred mouse strains display a variety of phenotypic differences. Many of these phenotypic traits mimic human conditions, making them excellent models in the study of complex phenotypes commonly observed in human health-related disoders. Gene expression variation among inbred mouse strains is often analyzed as a potential access point to genes responsible for this phenotypic variation. Here we describe the construction and application of a computational pipeline that identifies and returns a set of genes that display statistically significant differences in their expression between strains. Analyses were conducted on samples· that represent different sample size scenarios for validation of the pipeline, which require different statistical tests for significance (secondary oocyte for single sample per strain data and hippocampus for multiple sample per strain data). A 0.5 TPM threshold was used to address potential false positives due to sampling error. The threshold helped reduce the false positives by 6503 genes for the t-test and by 603 genes for the ANOVA. A confidence threshold of 99% (0.01) was used to test for significance. It was observed from the t-test that 7.7% of the genes displayed significant gene expression differences between strains in the secondary oocyte and 37.1 % of the genes displayed significant gene expression differences between strains in hippocampus. From the ANOVA and Scheffe test analysis, strains 129S1/SvlmJ and PWD/Ph have the greatest number of genes that show significant expression differences in hippocampus while A/J and C57BL/6J have the fewest. 333 genes were observed to be differentially expressed between all strains in hippocampus and were subjected to a molecular enrichment analysis.

Please contact the Joan Staats Library for information regarding this document.