Artificial Intelligence and Circulating Cell-Free DNA Methylation Profiling: Mechanism and Detection of Alzheimer's Disease.

Background: Despite extensive efforts, significant gaps remain in our understanding of Alzheimer's disease (AD) pathophysiology. Novel approaches using circulating cell-free DNA (cfDNA) have the potential to revolutionize our understanding of neurodegenerative disorders. Methods: We performed DNA methylation profiling of cfDNA from AD patients and compared them to cognitively normal controls. Six Artificial Intelligence (AI) platforms were utilized for the diagnosis of AD while enrichment analysis was used to elucidate the pathogenesis of AD. Results: A total of 3684 CpGs were significantly (adj. p-value < 0.05) differentially methylated in AD versus controls. All six AI algorithms achieved high predictive accuracy (AUC = 0.949−0.998) in an independent test group. As an example, Deep Learning (DL) achieved an AUC (95% CI) = 0.99 (0.95−1.0), with 94.5% sensitivity and specificity. Conclusion: We describe numerous epigenetically altered genes which were previously reported to be differentially expressed in the brain of AD sufferers. Genes identified by AI to be the best predictors of AD were either known to be expressed in the brain or have been previously linked to AD. We highlight enrichment in the Calcium signaling pathway, Glutamatergic synapse, Hedgehog signaling pathway, Axon guidance and Olfactory transduction in AD sufferers. To the best of our knowledge, this is the first reported genome-wide DNA methylation study using cfDNA to detect AD.

Urine metabolomic biomarkers for prediction of isolated fetal congenital heart defect.

OBJECTIVE: To identify maternal second and third trimester urine metabolomic biomarkers for the detection of fetal congenital heart defects (CHDs). STUDY DESIGN: This was a prospective study. Metabolomic analysis of randomly collected maternal urine was performed, comparing pregnancies with isolated, non-syndromic CHDs versus unaffected controls. Mass spectrometry (liquid chromatography and direct injection and tandem mass spectrometry, LC-MS-MS) as well as nuclear magnetic resonance spectrometry, 1H NMR, were used to perform the analyses between 14 0/7 and 37 0/7 weeks gestation. A total of 36 CHD cases and 41 controls were compared. Predictive algorithms using urine markers alone or combined with, clinical and ultrasound (US) (four-chamber view) predictors were developed and compared. RESULTS: A total of 222 metabolites were identified, of which 16 were overlapping between the two platforms. Twenty-three metabolite concentrations were found in significantly altered in CHD gestations on univariate analysis. The concentration of methionine was most significantly altered. A predictive algorithm combining metabolites (histamine, choline, glucose, formate, methionine, and carnitine) plus US four-chamber view achieved an AUC = 0.894; 95% CI, 0814-0.973 with a sensitivity of 83.8% and specificity of 87.8%. Enrichment pathway analysis identified several lipid related pathways that are dysregulated in CHD, including phospholipid biosynthesis, phosphatidylcholine biosynthesis, phosphatidylethanolamine biosynthesis, and fatty acid metabolism. This could be consistent with the increased risk of CHD in diabetic pregnancies. CONCLUSIONS: We report a novel, noninvasive approach, based on the analysis of maternal urine for isolated CHD detection. Further, the dysregulation of lipid- and folate metabolism appears to support prior data on the mechanism of CHD.