The coefficient of variance (CV) reflects the variance of gene expression levels in each library.
While the first PC explains the most variance in gene expression, lower PCs (exceeding our variance threshold) were retained in order to capture different aspects of cluster variance.
Because increased paternal age and autism were both negatively related to overall variance in gene expression, it was predicted that the overall variance in gene expression for children of older fathers would be similar to that in children with autism.
We hypothesized, therefore, that factors that modified risk for autism in the general population might alter the variance in gene expression in a manner that resembled the variance in gene expression in children with autism.
These marked differences in tissue composition constitute a major determinant of inter-sample variance in gene expression level for all genes unevenly expressed in the different structures.
As such, variance in gene expression leads to dramatic variation in glucosinolate structure and serves as evidence of genetic and biochemical neo-functionalization [56], [71], [72].
Despite a paucity of large expression changes, we found a strong correlation between the variance of gene expression change and species divergence (R2 = 0.90).
The first principal component (PC1) accounted for 39% of the overall variance in gene expression and second principal component (PC2) accounted for 36% of the variance.
To fit the assumption of equal variance of gene expression per group, genes displaying extreme variation (too small or too large) were excluded from the analysis.
We used the CV instead of the variance because in general there is a positive correlation between the mean and variance of gene expression levels.
