For one thing, not every case of replacement counts as substitution: the expression we replace with its synonym within a larger expression must be a constituent of the larger expression.
By treating Akata and Mutu-I cells with IgG and TGF β, compounds that mimic physiologically relevant mechanisms triggering the transition from latent to lytic infection, we found the autophagosomal marker LC3-II to increase at the early times, but decrease thereafter, in correlation with a larger expression of EBV early lytic genes.
To investigate whether each RP gene expression pattern was identical, we performed cluster analysis with a large gene expression dataset (3281 genes, see Additional file 2).
We constructed a constraints-based genome-scale stoichiometric model of S. coelicolor metabolism, based on earlier similar models [ 5, 6], and integrated the model predictions with a large gene expression dataset [ 7].
For example, a pair of genes that are significantly coordinated over 10 different biological conditions with a relatively small magnitude of expression change should be, according to our approach, 5-fold more coordinated than a pair of genes that are coordinated over two different biological perturbations with a relatively larger expression change.
We included G. tomentosum in this study because it had a greater number of homoeologous genes with large expression biases (i.e. a broader distribution of A/D expression ratios) than other polyploids.
The number of variants showing differential allelic expression was generally biased towards genes with high coverage and alleles with large expression differences between the treatments.
To focus on features with large expression differences, there were 276 (32% of differentially expressed genes) features with a minimum of a 3-fold change expression difference between the sexes in at least one comparison.
As an example of the kind of meta-analysis that is possible with a large public microarray expression database, we identify genes with very stable expression patterns, suitable as reference or control genes in northern blots, and genes with highly variable expression patterns in 162 Arabidopsis experiments.
Relative to the complete set described above, a smaller proportion of proteins from genes with large expression changes was found to have full-length homologues in non-mammalian species (with the largest difference in the fly).
