A Context-Specific, Literature-Supported Framework for Validating Stress Response Differentially Expressed Gene Sets

Computational models of stress responses identify genes underlying physiological adaptation, but their utility depends on rigorous validation. Often, gene activity reflects both adaptive mechanisms and noise. Here, we develop a framework that leverages public databases to support the subselection of biologically supported model genes for temperature-stress responses. We test our framework on a model that identified and categorized differentially expressed genes (DEGs) into Key-Response, Treatment-Specific, Noisy, and Support groups based on inter-individual gene expression variability before and after treatment. The first three groups were hypothesized to constitute a Principal Response. To validate these groupings, we constructed protein-protein interaction (PPI) networks using the Human Protein Atlas and STRING. The main contribution of this work is the implementation of second-order connections restricted to those made via DEGs, ensuring connectivity reflects condition-specific responses rather than generic hubs. Across two temperature conditions, >75% of Principal Response genes assembled into subnetworks of interactions significantly larger than random expectations. Support Group genes also showed strong interconnectivity and enrichment for housekeeping genes. STRING confirmed PPI enrichment but produced less stable results than our framework. By emphasizing DEG-restricted second-order connections, we address limitations of context-free enrichment methods and strengthen biological evaluation of computational models of differential gene expression. STATEMENT OF SIGNIFCANCE Computational models, old and new, are used to identify and highlight differentially expressed genes that work together to respond to certain conditions or phenotypes. Since gene expression and the statistical methods used to characterize it are inherently noisy, researchers’ models usually sub-select a group (or groups) of genes which are thought to be of elevated biological importance. However, outputted gene sets should be evaluated for biological ground-truth before they can be utilized further. Often, this biological validation requires experimental testing such as knockdown studies or pairwise epistasis analyses which may be burdensome in cost and time. Here, we present a database-powered framework for supporting the mechanistic plausibility of a subgroup of important DEGs using functional proteomic data. This simple but generalizable algorithm develops protein-protein interaction networks, which are known to be considerably reflective of genetic epistatic networks, that may be more specific to cellular contexts compared to existing methods. This provides researchers with a preliminary tool to test the biological plausibility of their model-selected genes in forming adaptive response mechanisms before they proceed to experimental validation. Competing Interest Statement The authors have declared no competing interest. Footnotes

has been rewritten to better explain the algorithm, emphasize its generalizability, and explain its utility and limitations. More findings have been added including comparing results to those from limma, a standardly used software. More specifics were added throughout, including in the methods section where a public repository of the source code can now be found. Citations added and further small revisions made throughout.