Predictors of healthy physiological aging across generations in a 30-year population-based cohort study: the Doetinchem Cohort Study.

BACKGROUND: Predicting healthy physiological aging is of major interest within public health research. However, longitudinal studies into predictors of healthy physiological aging that include numerous exposures from different domains (i.e. the exposome) are scarce. Our aim is to identify the most important exposome-related predictors of healthy physiological aging over the life course and across generations. METHODS: Data were used from 2815 participants from four generations (generation 1960s/1950s/1940s/1930s aged respectively 20-29/30-39/40-49/50-59 years old at baseline, wave 1) of the Doetinchem Cohort Study who were measured every 5 years for 30 years. The Healthy Aging Index, a physiological aging index consisting of blood pressure, glucose, creatinine, lung function, and cognitive functioning, was measured at age 46-85 years (wave 6). The average exposure and trend of exposure over time of demographic, lifestyle, environmental, and biological exposures were included, resulting in 86 exposures. Random forest was used to identify important predictors. RESULTS: The most important predictors of healthy physiological aging were overweight-related (BMI, waist circumference, waist/hip ratio) and cholesterol-related (using cholesterol lowering medication, HDL and total cholesterol) measures. Diet and educational level also ranked in the top of important exposures. No substantial differences were observed in the predictors of healthy physiological aging across generations. The final prediction model's performance was modest with an R2 of 17%. CONCLUSIONS: Taken together, our findings suggest that longitudinal cardiometabolic exposures (i.e. overweight- and cholesterol-related measures) are most important in predicting healthy physiological aging. This finding was similar across generations. More work is needed to confirm our findings in other study populations.

Predicting self-perceived general health status using machine learning: an external exposome study.

BACKGROUND: Self-perceived general health (SPGH) is a general health indicator commonly used in epidemiological research and is associated with a wide range of exposures from different domains. However, most studies on SPGH only investigated a limited set of exposures and did not take the entire external exposome into account. We aimed to develop predictive models for SPGH based on exposome datasets using machine learning techniques and identify the most important predictors of poor SPGH status. METHODS: Random forest (RF) was used on two datasets based on personal characteristics from the 2012 and 2016 editions of the Dutch national health survey, enriched with environmental and neighborhood characteristics. Model performance was determined using the area under the curve (AUC) score. The most important predictors were identified using a variable importance procedure and individual effects of exposures using partial dependence and accumulated local effect plots. The final 2012 dataset contained information on 199,840 individuals and 81 variables, whereas the final 2016 dataset had 244,557 individuals with 91 variables. RESULTS: Our RF models had overall good predictive performance (2012: AUC = 0.864 (CI: 0.852-0.876); 2016: AUC = 0.890 (CI: 0.883-0.896)) and the most important predictors were "Control of own life", "Physical activity", "Loneliness" and "Making ends meet". Subjects who felt insufficiently in control of their own life, scored high on the De Jong-Gierveld loneliness scale or had difficulty in making ends meet were more likely to have poor SPGH status, whereas increased physical activity per week reduced the probability of poor SPGH. We observed associations between some neighborhood and environmental characteristics, but these variables did not contribute to the overall predictive strength of the models. CONCLUSIONS: This study identified that within an external exposome dataset, the most important predictors for SPGH status are related to mental wellbeing, physical exercise, loneliness, and financial status.

Using random forest to identify longitudinal predictors of health in a 30-year cohort study.

Due to the wealth of exposome data from longitudinal cohort studies that is currently available, the need for methods to adequately analyze these data is growing. We propose an approach in which machine learning is used to identify longitudinal exposome-related predictors of health, and illustrate its potential through an application. Our application involves studying the relation between exposome and self-perceived health based on the 30-year running Doetinchem Cohort Study. Random Forest (RF) was used to identify the strongest predictors due to its favorable prediction performance in prior research. The relation between predictors and outcome was visualized with partial dependence and accumulated local effects plots. To facilitate interpretation, exposures were summarized by expressing them as the average exposure and average trend over time. The RF model's ability to discriminate poor from good self-perceived health was acceptable (Area-Under-the-Curve = 0.707). Nine exposures from different exposome-related domains were largely responsible for the model's performance, while 87 exposures seemed to contribute little to the performance. Our approach demonstrates that ML can be interpreted more than widely believed, and can be applied to identify important longitudinal predictors of health over the life course in studies with repeated measures of exposure. The approach is context-independent and broadly applicable.

Genomic analysis of European bovine Staphylococcus aureus from clinical versus subclinical mastitis.

Intramammary infections (IMI) with Staphylococcus aureus are a common cause of bovine mastitis and can result in both clinical (CM) or subclinical mastitis (SCM). Although bacterial isolates of S. aureus differ in their virulence potential it is largely unclear which bacterial virulence factors are responsible for increased clinical severity. We performed a genome wide association study and used a generalized linear mixed model to investigate the correlation between gene carriage, lineage and clinical outcome of IMI in a collection of S. aureus isolates from cattle with CM (n = 125) and SCM (n = 151) from 11 European countries. An additional aim was to describe the genetic variation of bovine S. aureus in Europa. The dominant lineages in our collection were clonal complex (CC) 151 (81/276, 29.3%), CC97 (54/276, 19.6%), CC479 (32/276, 11.6%) and CC398 (19/276, 6.9%). Virulence and antimicrobial resistance (AMR) gene carriage was highly associated with CC. Among a selection of nine virulence and AMR genes, CC151, CC479 and CC133 carried more virulence genes than other CCs, and CC398 was associated with AMR gene carriage. Whereas CC151, CC97 were widespread in Europe, CC479, CC398 and CC8 were only found in specific countries. Compared to CC151, CC479 was associated with CM rather than SCM (OR 3.62; 95% CI 1.38-9.50) and the other CCs were not. Multiple genes were associated with CM, but due to the clustering within CC of carriage of these genes, it was not possible to differentiate between the effect of gene carriage and CC on clinical outcome of IMI. Nevertheless, this study demonstrates that characterization of S. aureus CC and virulence genes helps to predict the likelihood of the occurrence of CM following S. aureus IMI and highlights the potential benefit of diagnostics tools to identify S. aureus CC during bovine mastitis.

Activation of a Bovine Mammary Epithelial Cell Line by Ruminant-Associated Staphylococcus aureus is Lineage Dependent.

Bovine mastitis is a costly disease to the dairy industry and intramammary infections (IMI) with Staphylococcus aureus are a major cause of mastitis. Staphylococcus aureus strains responsible for mastitis in cattle predominantly belong to ruminant-associated clonal complexes (CCs). Recognition of pathogens by bovine mammary epithelial cells (bMEC) plays a key role in activation of immune responsiveness during IMI. However, it is still largely unknown to what extent the bMEC response differs according to S. aureus CC. The aim of this study was to determine whether ruminant-associated S. aureus CCs differentially activate bMEC. For this purpose, the immortalized bMEC line PS was stimulated with S. aureus mastitis isolates belonging to four different clonal complexes (CCs; CC133, CC479, CC151 and CC425) and interleukin 8 (IL-8) release was measured as indicator of activation. To validate our bMEC model, we first stimulated PS cells with genetically modified S. aureus strains lacking (protein A, wall teichoic acid (WTA) synthesis) or expressing (capsular polysaccharide (CP) type 5 or type 8) factors expected to affect S. aureus recognition by bMEC. The absence of functional WTA synthesis increased IL-8 release by bMEC in response to bacterial stimulation compared to wildtype. In addition, bMEC released more IL-8 after stimulation with S. aureus expressing CP type 5 compared to CP type 8 or a strain lacking CP expression. Among the S. aureus lineages, isolates belonging to CC133 induced a significantly stronger IL-8 release from bMEC than isolates from the other CCs, and the IL-8 response to CC479 was higher compared to CC151 and CC425. Transcription levels of IL-8, tumor necrosis factor alpha (TNFα), serum amyloid A3 (SAA3), Toll-like receptor (TLR)-2 and nuclear factor κB (NF-κB) in bMEC after bacterial stimulation tended to follow a similar pattern as IL-8 release, but there were no significant differences between the CCs. This study demonstrates a differential activation of bMEC by ruminant-associated CCs of S. aureus, which may have implications for the severity of mastitis during IMI by S. aureus belonging to these lineages.

High Production of LukMF' in Staphylococcus aureus Field Strains Is Associated with Clinical Bovine Mastitis.

Staphylococcus aureus, a major cause of bovine mastitis, produces a wide range of immune-evasion molecules. The bi-component leukocidin LukMF' is a potent killer of bovine neutrophils in vitro. Since the role of LukMF' in development of bovine mastitis has not been studied in natural infections, we aimed to clarify whether presence of the lukM-lukF' genes and production levels of LukMF' are associated with clinical severity of the disease. Staphylococcus aureus was isolated from mastitis milk samples (38 clinical and 17 subclinical cases) from 33 different farms. The lukM-lukF' genes were present in 96% of the isolates. Remarkably, 22% of the lukM-lukF'-positive S. aureus isolates displayed a 10-fold higher in vitro LukMF' production than the average of the lower-producing ones. These high producing isolates were cultured significantly more frequently from clinical than subclinical mastitis cases. Also, the detection of LukM protein in milk samples was significantly associated with clinical mastitis and high production in vitro. The high producing LukMF' strains all belonged to the same genetic lineage, spa-type t543. Analysis of their global toxin gene regulators revealed a point mutation in the Repressor of toxins (rot) gene which results in a non-functional start codon, preventing translation of rot. This mutation was only identified in high LukMF' producing isolates and not in low LukMF' producing isolates. Since rot suppresses the expression of various toxins including leukocidins, this mutation is a possible explanation for increased LukMF' production. Identification of high LukMF' producing strains is of clinical relevance and can potentially be used as a prognostic marker for severity of mastitis.