Genome-wide transcriptome profiling identifies genes that are prone to differential expression across contexts ("common DEGs"), as well as specific changes relevant to the experimental manipulation. Distinguishing common DEGs from those that are specifically changed in a context of interest will allow more efficient inference of relevant mechanisms and a more systematic understanding of the biological process under scrutiny. Currently, common changes can only be identified through the laborious manual curation of highly controlled experiments, an inordinately time-consuming and impractical endeavor. Here we pioneer a method for identifying common patterns using generative neural networks. This method produces a background set of transcriptomic experiments from which a gene and pathway-specific null distribution can be generated. By comparing the set of differentially expressed genes found in a target experiment against the background set, common results can easily be separated from specific ones. This "Specific cOntext Pattern Highlighting In Expression data" (SOPHIE) method is broadly applicable to new platforms or any species with a large collection of unannotated gene expression data. We apply SOPHIE to diverse datasets including human, including human cancer, and bacterial datasets. SOPHIE recapitulates previously described common DEGs, and our molecular validation indicates it detects highly specific, but low magnitude, biologically relevant, transcriptional changes. SOPHIE's measure of specificity can complement log fold change activity generated from traditional differential expression analyses by, for example, filtering the set of changed genes to identify those that are specifically relevant to the experimental condition of interest. Consequently, these results can inform future research directions.