A supervised learning method for classifying methylation disorders.

BACKGROUND: DNA methylation is one of the most stable and well-characterized epigenetic alterations in humans. Accordingly, it has already found clinical utility as a molecular biomarker in a variety of disease contexts. Existing methods for clinical diagnosis of methylation-related disorders focus on outlier detection in a small number of CpG sites using standardized cutoffs which differentiate healthy from abnormal methylation levels. The standardized cutoff values used in these methods do not take into account methylation patterns which are known to differ between the sexes and with age. RESULTS: Here we profile genome-wide DNA methylation from blood samples drawn from within a cohort composed of healthy controls of different age and sex alongside patients with Prader-Willi syndrome (PWS), Beckwith-Wiedemann syndrome, Fragile-X syndrome, Angelman syndrome, and Silver-Russell syndrome. We propose a Generalized Additive Model to perform age and sex adjusted outlier analysis of around 700,000 CpG sites throughout the human genome. Utilizing z-scores among the cohort for each site, we deployed an ensemble based machine learning pipeline and achieved a combined prediction accuracy of 0.96 (Binomial 95% Confidence Interval 0.868[Formula: see text]0.995). CONCLUSION: We demonstrate a method for age and sex adjusted outlier detection of differentially methylated loci based on a large cohort of healthy individuals. We present a custom machine learning pipeline utilizing this outlier analysis to classify samples for potential methylation associated congenital disorders. These methods are able to achieve high accuracy when used with machine learning methods to classify abnormal methylation patterns.

Corrected QT Interval-Polygenic Risk Score and Its Contribution to Type 1, Type 2, and Type 3 Long-QT Syndrome in Probands and Genotype-Positive Family Members.

BACKGROUND: Long-QT syndrome (LQTS) is characterized by a prolonged heart rate-corrected QT interval (QTc). Genome-wide association studies identified common genetic variants that collectively explain ≈8% to 10% of QTc variation in the general population. METHODS: Overall, 423 patients with LQT1, LQT2, or LQT3 were genotyped for 61 QTc-associated genetic variants used in a prototype QTc-polygenic risk score (QTc-PRS). A weighted QTc-PRS (range, 0-154.8 ms) was calculated for each patient, and the FHS (Framingham Heart Study) population-based reference cohort (n=853). RESULTS: The average QTc-PRS in LQTS was 88.0±7.2 and explained only ≈2.0% of the QTc variability. The QTc-PRS in LQTS probands (n=137; 89.3±6.8) was significantly greater than both FHS controls (87.2±7.4, difference-in-means±SE: 2.1±0.7, P<0.002) and LQTS genotype-positive family members (87.5±7.4, difference-in-mean, 1.8±.7, P<0.009). There was no difference in QTc-PRS between symptomatic (n=156, 88.6±7.3) and asymptomatic patients (n=267; 87.7±7.2, difference-in-mean, 0.9±0.7, P=0.15). LQTS patients with a QTc≥480 ms (n=120) had a significantly higher QTc-PRS (89.3±6.7) than patients with a QTc<480 ms (n=303, 87.6±7.4, difference-in-mean, 1.7±0.8, P<0.05). There was no difference in QTc-PRS or QTc between genotypes. CONCLUSIONS: The QTc-PRS explained <2% of the QTc variability in our LQT1, LQT2, and LQT3 cohort, contributing 5× less to their QTc value than in the general population. This prototype QTc-PRS does not distinguish/predict the clinical outcomes of individuals with LQTS.