Machine learning methods have been used in identifying omics markers for a variety of phenotypes. We aimed to examine whether a supervised machine learning algorithm can improve identification of alcohol-associated transcriptomic markers. In this study, we analysed array-based, whole-blood derived expression data for 17 873 gene transcripts in 5508 Framingham Heart Study participants. By using the Boruta algorithm, a supervised random forest (RF)-based feature selection method, we selected twenty-five alcohol-associated transcripts. In a testing set (30 % of entire study participants), AUC (area under the receiver operating characteristics curve) of these twenty-five transcripts were 0·73, 0·69 and 0·66 for non-drinkers v. moderate drinkers, non-drinkers v. heavy drinkers and moderate drinkers v. heavy drinkers, respectively. The AUC of the selected transcripts by the Boruta method were comparable to those identified using conventional linear regression models, for example, AUC of 1958 transcripts identified by conventional linear regression models (false discovery rate < 0·2) were 0·74, 0·66 and 0·65, respectively. With Bonferroni correction for the twenty-five Boruta method-selected transcripts and three CVD risk factors (i.e. at P < 6·7e-4), we observed thirteen transcripts were associated with obesity, three transcripts with type 2 diabetes and one transcript with hypertension. For example, we observed that alcohol consumption was inversely associated with the expression of DOCK4, IL4R, and SORT1, and DOCK4 and SORT1 were positively associated with obesity, and IL4R was inversely associated with hypertension. In conclusion, using a supervised machine learning method, the RF-based Boruta algorithm, we identified novel alcohol-associated gene transcripts.

Conjoint analysis is widely used for estimating the effects of a large number of treatments on multidimensional decision-making. However, it is this substantive advantage that leads to a statistically undesirable property, multiple hypothesis testing. Existing applications of conjoint analysis except for a few do not correct for the number of hypotheses to be tested, and empirical guidance on the choice of multiple testing correction methods has not been provided. This paper first shows that even when none of the treatments has any effect, the standard analysis pipeline produces at least one statistically significant estimate of average marginal component effects in more than 90% of experimental trials. Then, we conduct a simulation study to compare three well-known methods for multiple testing correction, the Bonferroni correction, the Benjamini–Hochberg procedure, and the adaptive shrinkage (Ash). All three methods are more accurate in recovering the truth than the conventional analysis without correction. Moreover, the Ash method outperforms in avoiding false negatives, while reducing false positives similarly to the other methods. Finally, we show how conclusions drawn from empirical analysis may differ with and without correction by reanalyzing applications on public attitudes toward immigration and partner countries of trade agreements.

Healthy dietary patterns may protect against age-related cognitive decline, but results of studies have been inconsistent and few have had extensive longitudinal follow-up with comprehensive cognitive testing. The aim of the present study was to determine associations of dietary patterns with trajectories of global- and domain-specific cognitive change over a 12-year period. Data from 863 community-dwelling, dementia-free participants from the Lothian Birth Cohort 1936 study of ageing completed a FFQ at baseline (aged 70 years) and underwent cognitive testing at baseline, and at the ages of 73, 76, 79 and 82 years. Composite cognitive scores were constructed for four cognitive domains (visuospatial ability, processing speed, memory and verbal ability) and global cognitive function. A Mediterranean-style pattern and a traditional pattern were derived using principal component analysis of self-reported dietary intakes. In fully adjusted latent growth curve models, higher baseline adherence to the Mediterranean-style dietary pattern (β = 0·056, P = 0·009) and lower baseline adherence to the traditional dietary pattern (β  =  −0·087, P < 0·001) were cross-sectionally associated with better verbal ability. A slightly steeper decline in verbal ability over 12 years was observed in those with higher Mediterranean-style diet scores at baseline (β = −0·003, P = 0·008). All other associations were non-significant. Our findings in this well-characterised Scottish cohort indicate that adherence to a healthy Mediterranean-style diet is associated cross-sectionally with better verbal (crystallised) ability, with the converse being true for the traditional diet. A healthier baseline diet did not predict a reduced risk of global- or domain-specific cognitive decline.

CVD is the leading cause of death worldwide and, after dementia, is the second biggest cause of death for women. In England, it accounts for one in four of all deaths. Lifestyle modifications represent the primary route both to reduce CVD risk factors and prevent CVD outcomes. Diet constitutes one of the key modifiable risk factors in the aetiology of CVD. We investigated the relationship between nine main dietary indices and a comprehensive range of CVD risk factors in 2590 women from TwinsUK. After adjustment for multiple testing, we found that the Dietary Approaches to Stop Hypertension (DASH) diet was inversely correlated with some of the most common CVD risk factors (BMI, visceral fat (VF), TAG, insulin, homoeostasis model assessment of insulin resistance (HOMA2-IR) and atherosclerotic CVD (ASCVD) risk) with PFDR ranging from 6·28 × 10−7 to 5·63 × 10−4. Similar association patterns were detected across most of the dietary indices analysed. In our post hoc investigation, to determine if any specific food groups were driving associations between the DASH score and markers of cardiometabolic risk, we found that increased BMI, VF, HOMA2-IR, ASCVD risk, insulin and TAG levels were directly correlated with red meat consumption (PFDR ranging from 4·65 × 10−9 to 7·98 × 10−3) and inversely correlated with whole-grain cereal consumption (PFDR ranging from 1·26 × 10−6 to 8·28 × 10−3). Our findings revealed that the DASH diet is associated with a more favourable CVD risk profile, suggesting that this diet may be a candidate dietary pattern to supplement current UK dietary recommendations for CVD prevention.

The lithostratigraphic characteristics of the iconic Blue Lias Formation of southern Britain are influenced by sedimentation rates and stratigraphic gaps. Evidence for regular sedimentary cycles is reassessed using logs of magnetic susceptibility from four sites as an inverse proxy for carbonate content. Standard spectral analysis, including allowing for false discovery rates, demonstrates several scales of regular cyclicity in depth. Bayesian probability spectra provide independent confirmation of at least one scale of regular cyclicity at all sites. The frequency ratios between the different scales of cyclicity are consistent with astronomical forcing of climate at the periods of the short eccentricity, obliquity and precession cycles. Using local tuned time scales, 62 ammonite biohorizons have minimum durations of 0.7 to 276 ka, with 94% of them <41 ka. The duration of the Hettangian Stage is ≥2.9 Ma according to data from the West Somerset and Devon/Dorset coasts individually, increasing to ≥3.7 Ma when combined with data from Glamorgan and Warwickshire. A composite time scale, constructed using the tuned time scales plus correlated biohorizon limits treated as time lines, allows for the integration of local stratigraphic gaps. This approach yields an improved duration for the Hettangian Stage of ≥4.1 Ma, a figure that is about twice that suggested in recent time scales.

The absence of a dedicated transport for disaccharides in the intestine implicates that the metabolic use of dietary lactose relies on its prior hydrolysis at the intestinal brush border. Consequently, lactose in blood or urine has mostly been associated with specific cases in which the gastrointestinal barrier is damaged. On the other hand, lactose appears in the blood of lactating women and has been detected in the blood and urine of healthy men, indicating that the presence of lactose in the circulation of healthy subjects is not incompatible with normal physiology. In this cross-over study we have characterised the postprandial kinetics of lactose, and its major constituent, galactose, in the serum of fourteen healthy men who consumed a unique dose of 800 g milk or yogurt. Genetic testing for lactase persistence and microbiota profiling of the subjects were also performed. Data revealed that lactose does appear in serum after dairy intake, although with delayed kinetics compared with galactose. Median serum concentrations of approximately 0·02 mmol/l lactose and approximately 0·2 mmol/l galactose were observed after the ingestion of milk and yogurt respectively. The serum concentrations of lactose were inversely correlated with the concentrations of galactose, and the variability observed between the subjects’ responses could not be explained by the presence of the lactase persistence allele. Finally, lactose levels have been associated with the abundance of the Veillonella genus in faecal microbiota. The measurement of systemic lactose following dietary intake could provide information about lactose metabolism and nutrient transport processes under normal or pathological conditions.

In a multiple testing context, we consider a semiparametric mixture model with two components where one component is known and corresponds to the distribution of p-values under the null hypothesis and the other component f is nonparametric and stands for the distribution under the alternative hypothesis. Motivated by the issue of local false discovery rate estimation, we focus here on the estimation of the nonparametric unknown component f in the mixture, relying on a preliminary estimator of the unknown proportion θ of true null hypotheses. We propose and study the asymptotic properties of two different estimators for this unknown component. The first estimator is a randomly weighted kernel estimator. We establish an upper bound for its pointwise quadratic risk, exhibiting the classical nonparametric rate of convergence over a class of Hölder densities. To our knowledge, this is the first result establishing convergence as well as corresponding rate for the estimation of the unknown component in this nonparametric mixture. The second estimator is a maximum smoothed likelihood estimator. It is computed through an iterative algorithm, for which we establish a descent property. In addition, these estimators are used in a multiple testing procedure in order to estimate the local false discovery rate. Their respective performances are then compared on synthetic data.

Statistical analysis is increasingly used in seed germination/viability studies across different disciplines. The objective of this opinion piece is to assess current trends in statistical analysis of such data, and draw attention of readers to the limitations of the usual inferential statistics in controlling error rates. The assessments are based on a survey of 429 papers published in 139 peer-reviewed journals in the past 11 years. My intention is to identify areas of concern across a wide range of studies. Accordingly, the areas of greatest concern found in the analysis of percentage seed germination and viability data were: (1) pseudoreplication and/or use of a few replicates; (2) ignoring assumptions of ANOVA and non-parametric tests (NPARTs); (3) uncritical data transformation; (4) arbitrary choice of multiple comparison tests; and (5) lack of emphasis on effect sizes. Given the prevalence of these problems, in my opinion we would be building a body of knowledge on a shaky ground. The discussions that follow will: (1) describe situations where germination data violate assumptions of ANOVA and NPARTs; (2) highlight the implications of the various problems to Type I and Type II error rates; and (3) indicate remedial measures based on the recent statistical literature.