High-throughput metabolomics investigations, when conducted in large human cohorts, represent a potentially powerful tool for elucidating the biochemical diversity underlying human health and disease. Large-scale metabolomics data sources, generated using either targeted or nontargeted platforms, are becoming more common. Appropriate statistical analysis of these complex high-dimensional data will be critical for extracting meaningful results from such large-scale human metabolomics studies. Therefore, we consider the statistical analytical approaches that have been employed in prior human metabolomics studies. Based on the lessons learned and collective experience to date in the field, we offer a step-by-step framework for pursuing statistical analyses of cohort-based human metabolomics data, with a focus on feature selection. We discuss the range of options and approaches that may be employed at each stage of data management, analysis, and interpretation and offer guidance on the analytical decisions that need to be considered over the course of implementing a data analysis workflow. Certain pervasive analytical challenges facing the field warrant ongoing focused research. Addressing these challenges, particularly those related to analyzing human metabolomics data, will allow for more standardization of as well as advances in how research in the field is practiced. In turn, such major analytical advances will lead to substantial improvements in the overall contributions of human metabolomics investigations.

No abstract

Background: Echocardiographic deformation-based ratios and novel multi-parametric scores have been suggested to discriminate transthyretin cardiac amyloidosis (ATTR-CM) from other causes of increased left ventricular wall thickness among patients referred for ATTR-CM evaluation. Their relative predictive accuracy has not been well studied. We sought to (1) identify echocardiographic parameters predictive of ATTR-CM and (2) compare the diagnostic accuracy of these parameters in patients with suspected ATTR-CM referred for technetium-99m-pyrophosphate scintigraphy. Methods: Echocardiograms from 598 patients referred to 3 major amyloidosis centers for technetium-99m-pyrophosphate to detect ATTR-CM were analyzed, including longitudinal strain (LS) analysis. Deformation ratios (septal apex to base ratio, relative apical sparing, ejection fraction to global LS), a multi-center European increased wall thickness score, and Mayo Clinic derived ATTR score (transthyretin cardiac amyloidosis score) were calculated. A logistic regression model was used to identify the parameters that best associated with a diagnosis of ATTR-CM. Comparison of the diagnostic capacity of the parameters was performed by receiver operating characteristic curves and the area under the curve (AUC). Results: Over half of the subjects (54.2%) were diagnosed with ATTR-CM (78% were men, median age of 76 years). Age, inferolateral wall thickness, and basal LS were the strongest predictors of ATTR-CM, AUC of 0.87 (95% CI: 0.83, 0.90), superior to the increased wall thickness score AUC of 0.78 (95% CI: 0.73, 0.83; P =0.004). An inferolateral wall thickness of ≥14 mm (AUC: 0.73) was as accurate as the published cut-offs for transthyretin cardiac amyloidosis score and septal apex to base (AUC: 0.72 and 0.69, P =0.8 and P =0.1, respectively), and was superior to ejection fraction to global LS and relative apical sparing (AUC: 0.64 and 0.53, P <0.001, respectively). A cut-off of ≥−8% for average basal LS (AUC: 0.76, CI: 0.72–0.79) had a similar area under the curve to transthyretin cardiac amyloidosis score (TCAS) ( P =0.2); outperforming the other indices ( P <0.01). Conclusion: Inferolateral wall thickness and average basal LS performed as well as or better than more complex echo ratios and multiparametric scores to predict ATTR-CM.