Background The identification of gene–gene and gene–environment interactions in genome-wide association studies is challenging due to the unknown nature of the interactions and the overwhelmingly large number of possible combinations. Parametric regression models are suitable to look for prespecified interactions. Nonparametric models such as tree ensemble models, with the ability to detect any unspecified interaction, have previously been difficult to interpret. However, with the development of methods for model explainability, it is now possible to interpret tree ensemble models efficiently and with a strong theoretical basis. Results We propose a tree ensemble- and SHAP-based method for identifying as well as interpreting potential gene–gene and gene–environment interactions on large-scale biobank data. A set of independent cross-validation runs are used to implicitly investigate the whole genome. We apply and evaluate the method using data from the UK Biobank with obesity as the phenotype. The results are in line with previous research on obesity as we identify top SNPs previously associated with obesity. We further demonstrate how to interpret and visualize interaction candidates. Conclusions The new method identifies interaction candidates otherwise not detected with parametric regression models. However, further research is needed to evaluate the uncertainties of these candidates. The method can be applied to large-scale biobanks with high-dimensional data.

BackgroundDengue and West Nile viruses are highly cross-reactive and have numerous parallels in geography, potential vector host (Aedes family of mosquitoes), and initial symptoms of infection. While the vast majority (> 80%) of both dengue and West Nile virus infections result in asymptomatic infections, a minority of individuals experience symptomatic infection and an even smaller proportion develop severe disease. The mechanisms by which these infections lead to severe disease in a subset of infected individuals is incompletely understood, but individual host differences including genetic factors and immune responses have been proposed. We sought to identify genetic risk factors that are associated with more severe disease outcomes for both viruses in order to shed light on possible shared mechanisms of resistance and potential therapeutic interventions.MethodsWe applied a search strategy using four major databases (Medline, PubMed, Embase, and Global Health) to find all known genetic associations identified to date with dengue or West Nile virus disease. Here we present a review of our findings and a meta-analysis of genetic variants identified.ResultsWe found genetic variations that are significantly associated with infections of these viruses. In particular we found variation within the OAS1 (meta-OR = 0.83, 95% CI: 0.69–1.00) and CCR5 (meta-OR = 1.29, 95% CI: 1.08–1.53) genes is significantly associated with West Nile virus disease, while variation within MICB (meta-OR = 2.35, 95% CI: 1.68–3.29), PLCE1 (meta-OR = 0.55, 95% CI: 0.42–0.71), MBL2 (meta-OR = 1.54, 95% CI: 1.02–2.31), and IFN-γ (meta-OR = 2.48, 95% CI: 1.30–4.71), is associated with dengue disease.ConclusionsDespite substantial heterogeneity in populations studied, genes examined, and methodology, significant associations with genetic variants were found across studies within both diseases. These gene associations suggest a key role for immune mechanisms in susceptibility to severe disease. Further research is needed to elucidate the role of these genes in disease pathogenesis and may reveal additional genetic factors associated with disease severity.Electronic supplementary materialThe online version of this article (10.1186/s12879-018-3186-6) contains supplementary material, which is available to authorized users.

The genus Flavivirus contains many mosquito-borne human pathogens of global epidemiological importance such as dengue virus, West Nile virus, and Zika virus, which has recently emerged at epidemic levels. Infections with these viruses result in divergent clinical outcomes ranging from asymptomatic to fatal. Myriad factors influence infection severity including exposure, immune status and pathogen/host genetics. Furthermore, pre-existing infection may skew immune pathways or divert immune resources. We profiled immune cells from dengue virus-infected individuals by multiparameter mass cytometry (CyTOF) to define functional status. Elevations in IFNβ were noted in acute patients across the majority of cell types and were statistically elevated in 31 of 36 cell subsets. We quantified response to in vitro (re)infection with dengue or Zika viruses and detected a striking pattern of upregulation of responses to Zika infection by innate cell types which was not noted in response to dengue virus. Significance was discovered by statistical analysis as well as a neural network-based clustering approach which identified unusual cell subsets overlooked by conventional manual gating. Of public health importance, patient cells showed significant enrichment of innate cell responses to Zika virus indicating an intact and robust anti-Zika response despite the concurrent dengue infection.

Background Blood smear microscopy remains the gold-standard method to diagnose and quantify malaria parasite density. In addition, parasite genotyping of select loci is the most utilized method for distinguishing recrudescent and new infections and to determine the number of strains per sample. In research settings, blood may be obtained from capillary or venous compartments, and results from these matrices have been used interchangeably. Our aim was to compare quantitative results for parasite density and strain complexity from both compartments. Methods In a prospective observational study, children and adults presenting with uncomplicated Plasmodium falciparum malaria, simultaneous capillary and venous blood smears and dried blood spots were collected over 42-days following treatment with artemether-lumefantrine. Blood smears were read by two microscopists, any discrepancies resolved by a third reader. Parasite DNA fingerprinting was conducted using six microsatellites. Bland Altman analysis and paired t-test/McNemar’s test were used to assess the difference in density readings and measurements. Results Two hundred twenty-three participants were included in the analysis (177 children (35 HIV-infected/142 HIV-uninfected), 21 HIV-uninfected pregnant women, and 25 HIV-uninfected non-pregnant adults). Parasite density measurements did not statistically differ between capillary and venous blood smears at the time of presentation, nor over the course of 42-day follow-up. Characterization of merozoite surface protein-2 (MSP-2) genetic polymorphism demonstrated a higher level of strain diversity at the time of presentation in venous samples, as compared with capillary specimens ( p = 0.02). There was a high degree of variability in genotype-corrected outcomes when pairs of samples from each compartment were compared using MSP-2 alone, although the variability was reduced with the use of multiple markers. Conclusions Parasite density measurements do not statistically differ between capillary and venous compartments in all studied demographic groups at the time of presentation with malaria, or over the course of follow-up. More strains were detected by MSP-2 genotyping in venous samples than in capillary samples at the time of malaria diagnosis. The use of multiple polymorphic markers reduces the impact of variability in strain detection on genotype-corrected outcomes. This study confirms that both capillary and venous compartments can be used for sampling with confidence in the clinical research setting. Trial registration The trial was registered at ClinicalTrials.gov under registration no. NCT01717885 . Electronic supplementary material The online version of this article (10.1186/s12879-019-4174-1) contains supplementary material, which is available to authorized users.

This descriptive study used both quantitative and qualitative research methods to examine the food and nutrient intake, food purchasing patterns and budgeting strategies of 29 sole‐parent low‐income families with dependent children living in Corio Shire, Victoria, in 1989–90. Expenditure on food and nonalcoholic beverages when compared with the average for all Australian households showed that the study families allocated a greater proportion, but similar amounts of money to cereals, dairy products, fruit and vegetables and miscellaneous foods, and less to meat, nonalcoholic beverages and food eaten away from home. Despite large differences in the amount of money spent in the first and second weeks of a social security payment period, the nutrient density of the parents' diets was maintained at a similar level in both weeks. The only exception was vitamin C, for which the median intake was significantly lower in the second week, consistent with reduced purchases of fruit and vegetables in the same period. In interviews with 13 of the parents a variety of purchasing and budgeting strategies were described, indicating common concerns with making the most of their limited resources. The data from this study support the notion that low‐income families give priority to food purchases above other expenses such as recreation. They also support the view that low‐income families are concerned about health and nutrition and manage to consume nutritionally sound diets under difficult circumstances. Nutrition education programs could support them in this endeavour.