Fusing Data Mining,Machine Learning and Traditional Statistics to Detect Biomarkers Associated with Depression

Background

Atheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study.

Methods

The study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009–2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators.

Results

After the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001).

Conclusion

The systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin.     

Climate‐driven changes to the spatio‐temporal distribution of the parasitic nematode,Haemonchus contortus,in sheep in Europe

Recent climate change has resulted in changes to the phenology and distribution of invertebrates worldwide. Where invertebrates are associated with disease, climate variability and changes in climate may also affect the spatio‐temporal dynamics of disease. Due to its significant impact on sheep production and welfare, the recent increase in diagnoses of ovine haemonchosis caused by the nematode Haemonchus contortus in some temperate regions is particularly concerning. This study is the first to evaluate the impact of climate change on H. contortus at a continental scale. A model of the basic reproductive quotient of macroparasites, Q₀, adapted to H. contortus and extended to incorporate environmental stochasticity and parasite behaviour, was used to simulate Pan‐European spatio‐temporal changes in H. contortus infection pressure under scenarios of climate change. Baseline Q₀ simulations, using historic climate observations, reflected the current distribution of H. contortus in Europe. In northern Europe, the distribution of H. contortus is currently limited by temperatures falling below the development threshold during the winter months and within‐host arrested development is necessary for population persistence over winter. In southern Europe, H. contortus infection pressure is limited during the summer months by increased temperature and decreased moisture. Compared with this baseline, Q₀ simulations driven by a climate model ensemble predicted an increase in H. contortus infection pressure by the 2080s. In northern Europe, a temporal range expansion was predicted as the mean period of transmission increased by 2–3 months. A bimodal seasonal pattern of infection pressure, similar to that currently observed in southern Europe, emerges in northern Europe due to increasing summer temperatures and decreasing moisture. The predicted patterns of change could alter the epidemiology of H. contortus in Europe, affect the future sustainability of contemporary control strategies, and potentially drive local adaptation to climate change in parasite populations.     

Hydrogen isotopic differences between C3 and C4 land plant lipids: consequences of compartmentation in C4 photosynthetic chemistry and C3 photorespiration

The ²H/¹H ratio of carbon‐bound H in biolipids holds potential for probing plant lipid biosynthesis and metabolism. The biochemical mechanism underlying the isotopic differences between lipids from C₃ and C₄ plants is still poorly understood. GC‐pyrolysis‐IRMS (gas chromatography‐pyrolysis‐isotope ratio mass spectrometry) measurement of the ²H/¹H ratio of leaf lipids from controlled and field grown plants indicates that the biochemical isotopic fractionation (ε²H_{lipid_biochem}) differed between C₃ and C₄ plants in a pathway‐dependent manner: ε²H_C4 > ε²H_C3 for the acetogenic pathway, ε²H_C4 < ε²H_C3 for the mevalonic acid pathway and the 1‐deoxy‐D‐xylulose 5‐phosphate pathway across all species examined. It is proposed that compartmentation of photosynthetic CO₂ fixation into C₄ mesophyll (M) and bundle sheath (BS) cells and suppression of photorespiration in C₄ M and BS cells both result in C₄ M chloroplastic pyruvate – the precursor for acetogenic pathway – being more depleted in ²H relative to pyruvate in C₃ cells. In addition, compartmentation in C₄ plants also results in (i) the transferable H of NADPH being enriched in ²H in C₄ M chloroplasts compared with that in C₃ chloroplasts for the 1‐deoxy‐D‐xylulose 5‐phosphate pathway pathway and (ii) pyruvate relatively ²H‐enriched being used for the mevalonic acid pathway in the cytosol of BS cells in comparison with that in C₃ cells.     

Aberrant DNA Methylation: Implications in Racial Health Disparity

BackgroundIncidence and mortality rates of colorectal carcinoma (CRC) are higher in African Americans (AAs) than in Caucasian Americans (CAs). Deficient micronutrient intake due to dietary restrictions in racial/ethnic populations can alter genetic and molecular profiles leading to dysregulated methylation patterns and the inheritance of somatic to germline mutations.ResultsDNA from the tumor of AA CRC patients, compared to adjacent normal tissues, contained 1,588 hypermethylated and 100 hypomethylated differentially methylated regions (DMRs). Whereas, 109 hypermethylated and 4 hypomethylated DMRs were observed in DNA from the tumor of CA CRC patients; representing a 14.6-fold and 25-fold change, respectively. Specifically; CHL1, 4 anti-inflammatory genes (i.e., NELL1, GDF1, ARHGEF4, and ITGA4), and 7 miRNAs (of which miR-9-3p and miR-124-3p have been implicated in CRC) were hypermethylated in DNA samples from AA patients with CRC. From the same sample set, RNAseq analysis revealed 108 downregulated genes (including 14 ribosomal proteins) and 34 upregulated genes (including POLR2B and CYP1B1 [targets of miR-124-3p]) in AA patients with CRC versus CA patients.ConclusionDNA methylation profile and/or products of its downstream targets could serve as biomarker(s) addressing racial health disparity.     

FLEXc: protein flexibility prediction using context-based statistics,predicted structural features,and sequence information

Background

The fluctuation of atoms around their average positions in protein structures provides important information regarding protein dynamics. This flexibility of protein structures is associated with various biological processes. Predicting flexibility of residues from protein sequences is significant for analyzing the dynamic properties of proteins which will be helpful in predicting their functions.

Results

In this paper, an approach of improving the accuracy of protein flexibility prediction is introduced. A neural network method for predicting flexibility in 3 states is implemented. The method incorporates sequence and evolutionary information, context-based scores, predicted secondary structures and solvent accessibility, and amino acid properties. Context-based statistical scores are derived, using the mean-field potentials approach, for describing the different preferences of protein residues in flexibility states taking into consideration their amino acid context.The 7-fold cross validated accuracy reached 61 % when context-based scores and predicted structural states are incorporated in the training process of the flexibility predictor.

Conclusions

Incorporating context-based statistical scores with predicted structural states are important features to improve the performance of predicting protein flexibility, as shown by our computational results. Our prediction method is implemented as web service called “FLEXc” and available online at: http://hpcr.cs.odu.edu/flexc.

     

Pollen—Tiny and ephemeral but not forgotten: New ideas on their ecology and evolution

Ecologists and evolutionary biologists have been interested in the functional biology of pollen since the discovery in the 1800s that pollen grains encompass tiny plants (male gametophytes) that develop and produce sperm cells. After the discovery of double fertilization in flowering plants, botanists in the early 1900s were quick to explore the effects of temperature and maternal nutrients on pollen performance, while evolutionary biologists began studying the nature of haploid selection and pollen competition. A series of technical and theoretic developments have subsequently, but usually separately, expanded our knowledge of the nature of pollen performance and how it evolves. Today, there is a tremendous diversity of interests that touch on pollen performance, ranging from the ecological setting on the stigma, structural and physiological aspects of pollen germination and tube growth, the form of pollen competition and its role in sexual selection in plants, virus transmission, mating system evolution, and inbreeding depression. Given the explosion of technical knowledge of pollen cell biology, computer modeling, and new methods to deal with diversity in a phylogenetic context, we are now more than ever poised for a new era of research that includes complex functional traits that limit or enhance the evolution of these deceptively simple organisms.