Subramanian, N. A. | Search bioRxiv

bioRxiv2025

Uncertainty-Aware Gene Rankings Reveal Key Players in Coexpression Networks

MotivationKey genes of a biological system are often prioritized by computing network science measures on a coexpression network inferred from transcriptomic data. But human population heterogeneity and modest sample sizes introduce uncertainty in the inferred coexpression network. Earlier studies have estimated this uncertainty using bootstrap resampling or similar approaches, but fewer have investigated how it propagates to downstream network analyses and affects gene prioritization. Methods and ResultsWe present a systematic workflow to propagate network uncertainty to downstream measures such as degree/PageRank centrality, with the goal of producing robust gene scorings/rankings. We specifically propose uncertainty-aware scorings, BooNS and BPNS, which utilize the spread of a centrality measure across bootstrapped coexpression networks to prioritize stable central genes. Across several (semi-)simulated and real-world (GTEx) datasets, BooNS and BPNS recover reference or tissue-specific genes significantly better than other traditional centrality rankings. This performance gap highlights the long-overdue adoption of uncertainty-aware gene ranking for stable biological inference. Availability and ImplementationCode and supplemental data are available at https://github.com/BIRDSgroup/Bootstrap-based_Node_Scorings_BNS, and https://tinyurl.com/BNS-suppl-data respectively.

bioinformatics↗

bioRxiv2023

Systems analysis of multiple diabetes-helminth cohorts reveals markers of disease-disease interaction

Understanding how the molecules in our body respond to the co-occurrence of two diseases in an individual (comorbidity) could lead to mechanistic insights into novel treatments for comorbid conditions. Studies have shown for instance that responses of our immune system to comorbid conditions could be more complex than the union of immune responses to each disease occurring separately, but a data-driven quantification of this complexity is lacking. In this study, we present a systematic methodology to quantify the interaction effect of two diseases on marker variables of interest (using a chronic inflammatory disease diabetes and parasitic infection helminth as illustrative disease pairs to identify cytokines or other immune markers that respond distinctively under a comorbid condition). To perform this systematic comorbidity analysis, we (i) collected and preprocessed data measurements from multiple single- and double-disease cohorts, (ii) extended differential expression analysis of such data to identify disease-disease interaction (DDI) markers (such as cytokines that respond antagonistically or synergistically to the double-disease condition relative to single-disease states), and (iii) interpreted the resulting DDI markers in the context of prior cytokine/immune-cell knowledgebases. We applied this three-step DDI methodology to multiple cohorts of helminth and diabetes (specifically, helminth-infected and helminth-treated individuals in diabetic and non-diabetic conditions, and non-disease control individuals), and identified cytokines such as IFN-{gamma}, TNF-, and IL-2 to be DDI markers acting at the interface of both diseases in data collected prior to helminth treatment. Validating our expectations, for these cytokines and other T helper Th-2 cytokines like IL-13 and IL-4, their DDI statuses were lost after treatment for helminth infection. For instance, the relative contribution of the DDI term in explaining the individual-to-individual variation of IFN-{gamma} and TNF- cytokines were 67.68% and 48.88% respectively before anthelmintics treatment and dropped to 6.09% and 14.56% respectively after treatment. Furthermore, signaling pathways like IL-10 and IL-4/IL-13 were found to be significantly enriched for genes targeted by certain DDI markers, thereby suggesting mechanistic hypotheses on how these DDI markers influence both diseases. Our results quantified the extent of helminth-diabetes DDI exhibited by various tested cytokine markers, and thereby delineated their role in the pathogenesis of both diseases. These results are promising and encourage the application of our DDI methodology (https://github.com/BIRDSgroup/DDI) to dissect the interaction between any two diseases, provided multi-cohort measurements of markers are available.

systems biology↗