Despite advances in cancer treatment, the 5-year mortality rate for oral cancers (OC) is 40%, mainly due to the lack of early diagnostics. To advance early diagnostics for high-risk and average-risk populations, we developed and evaluated machine-learning (ML) classifiers using metatranscriptomic data from saliva samples (n = 433) collected from oral premalignant disorders (OPMD), OC patients (n = 71) and normal controls (n = 171). Our diagnostic classifiers yielded a receiver operating characteristics (ROC) area under the curve (AUC) up to 0.9, sensitivity up to 83% (92.3% for stage 1 cancer) and specificity up to 97.9%. Our metatranscriptomic signature incorporates both taxonomic and functional microbiome features, and reveals a number of taxa and functional pathways associated with OC. We demonstrate the potential clinical utility of an AI/ML model for diagnosing OC early, opening a new era of non-invasive diagnostics, enabling early intervention and improved patient outcomes.
To prevent and treat chronic diseases, including cancer, a global application of systems biology is needed. We report here a whole blood transcriptome test that needs only 50 μl of capillary (fingerprick) blood. This test is suitable for global applications because the samples are preserved at ambient temperature for up to 4 weeks and the RNA preservative inactivates all pathogens, enabling safe transportation. Both the laboratory and bioinformatic steps are automated and performed in a clinical lab, which minimizes batch effects and creates unbiased datasets. Given its clinical testing performance and accessibility to traditionally underrepresented and diverse populations, this test offers a unique ability to reveal molecular mechanisms of disease and enable longitudinal, population-scale studies.
Limiting postprandial glycemic response (PPGR) is an important intervention in reducing the risk of chronic metabolic diseases and has been shown to impart significant health benefits in people with elevated levels of blood sugar. In this study, we collected gut microbiome activity data by assessing the metatranscriptome, and we measured the glycemic responses of 550 adults who consumed more than 30,000 meals, collectively, from omnivore or vegetarian/gluten-free diets. We demonstrate that gut microbiome activity, anthropometric factors, and food macronutrients modulate individual variation in glycemic response. We employ two predictive models, including a mixed-effects linear regression model ( R = 0.77) and a gradient boosting machine model ( R train = 0.80/ R 2 train = 0.64; R test = 0.64/ R 2 test = 0.40), which demonstrate variation in PPGR between individuals when ingesting the same foods. All features in the final mixed-effects linear regression model were significant ( p < 0.05) except for two features which were retained as suggestive: glutamine production pathways ( p = 0.08) and the interaction between tyrosine metabolizers and carbs ( p = 0.06). We introduce molecular functions as features in these two models, aggregated from microbial activity data, and show their statistically significant contributions to glycemic control. In summary, we demonstrate for the first time that metatranscriptomic activity of the gut microbiome is correlated with PPGR among adults.
Limiting post-meal glycemic response is an important factor in reducing the risk of chronic metabolic diseases, and contributes to significant health benefits in people with elevated levels of blood sugar. In this study, we collected gut microbiome activity (i.e., metatranscriptomic) data and measured the glycemic responses of 550 adults who consumed more than 27,000 meals from omnivore or vegetarian/gluten-free diets. We demonstrate that gut microbiome activity makes a statistically significant contribution to individual variation in glycemic response, in addition to anthropometric factors and the nutritional composition of foods. We describe a predictive model (multilevel mixed-effects regression) of variation in glycemic response among individuals ingesting the same foods. We introduce functional features aggregated from microbial activity data as candidates for association with mechanisms of glycemic control. In summary, we demonstrate for the first time that metatranscriptomic activity of the gut microbiome is correlated with glycemic response among adults. Figure 1, we recruited 550 adults (~66% female), and tracked their food intake, sleep, activity, and glycemic response for up to 2 weeks. 400 participants were Caucasian, and of the remaining 150 participants, 37% were Asian, 33% were Hispanic, and 30% were *This study was performed while all authors were at Viome Inc. METHODS As described in
Chronic diseases are the leading cause of morbidity and mortality globally. Yet, the majority of them have unknown etiologies, and genetic contribution is weak. In addition, many of the chronic diseases go through the cycles of relapse and remission, during which the genomic DNA does not change. This strongly suggests that human gene expression is the main driver of chronic disease onset and relapses. To identify the etiology of chronic diseases and develop more effective preventative measures, a comprehensive gene expression analysis of the human body is needed. Blood tissue is easy to access and contains a large number of expressed genes involved in many fundamental aspects of our physiology.We report here the development of a whole blood transcriptome clinical test that is high throughput, automated, inexpensive, and clinically validated. The test requires only 50 microliters of blood from a finger prick, enabling access by diverse populations that have been traditionally underrepresented in clinical research. The transcripts in the samples are preserved at the time of collection and can be stored and/or transported at ambient temperatures for up to 28 days. The sample preservative protects integrity, while also inactivating all pathogens (bacteria, fungi, and viruses), enabling safe transportation globally. Given its unique set of usability features and clinical performance, this test should be integrated into longitudinal, populationscale, systems biology studies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.