Search

Abstract Background Celiac disease is an autoimmune disorder triggered by dietary gluten in genetically predisposed individuals that primarily affects the small intestine. Studies have reported differentially abundant bacterial taxa in the gut microbiota of celiac patients compared with non-celiac controls. However, findings across studies have inconsistencies and no microbial signature of celiac disease has been defined so far. Results Here, we showed, by comparing celiac patients with their non-celiac 1st-degree relatives, that bacterial communities of related individuals have similar species occurrence and abundance compared with non-relatives, regardless the disease status. We also found in celiac patients a loss of bacterial species associated with fiber degradation, and host metabolic and immune modulation, as ruminiclostridia, ruminococci, Prevotella, and Akkermansia muciniphila species. We demonstrated that the differential abundance of bacterial species correlates to different dietary patterns observed between the two groups. For instance, Ruminiclostridium siraeum, Ruminococcus bicirculans, and Bacteroides plebeious, recognized as fiber-degraders, appear more abundant in non-celiac 1st-degree relatives, which have a vegetable consumption pattern higher than celiac patients. Pattern of servings per day also suggests a possible link between these species’ abundance and daily calorie intake. Conclusions Overall, we evidenced that a kinship approach could be valuable in unveiling potential celiac disease microbial traits, as well as the significance of dietary factors in shaping microbial profiles and their influence on disease development and progression. Our results pave the way for designing and adopting novel dietary strategies based on gluten-free fiber-enriched ingredients to improve disease management and patients' quality of life.

Abstract Background Machine learning (ML) has made a significant impact in medicine and cancer research; however, its impact in these areas has been undeniably slower and more limited than in other application domains. A major reason for this has been the lack of availability of patient data to the broader ML research community, in large part due to patient privacy protection concerns. High-quality, realistic, synthetic datasets can be leveraged to accelerate methodological developments in medicine. By and large, medical data is high dimensional and often categorical. These characteristics pose multiple modeling challenges. Methods In this paper, we evaluate three classes of synthetic data generation approaches; probabilistic models, classification-based imputation models, and generative adversarial neural networks. Metrics for evaluating the quality of the generated synthetic datasets are presented and discussed. Results While the results and discussions are broadly applicable to medical data, for demonstration purposes we generate synthetic datasets for cancer based on the publicly available cancer registry data from the Surveillance Epidemiology and End Results (SEER) program. Specifically, our cohort consists of breast, respiratory, and non-solid cancer cases diagnosed between 2010 and 2015, which includes over 360,000 individual cases. Conclusions We discuss the trade-offs of the different methods and metrics, providing guidance on considerations for the generation and usage of medical synthetic data.