How to increase our belief in discovered statistical interactions via large-scale association studies?

The understanding that differences in biological epistasis may impact disease risk, diagnosis, or disease management stands in wide contrast to the unavailability of widely accepted large-scale epistasis analysis protocols. Several choices in the analysis workflow will impact false-positive and false-negative rates. One of these choices relates to the exploitation of particular modelling or testing strategies. The strengths and limitations of these need to be well understood, as well as the contexts in which these hold. This will contribute to determining the potentially complementary value of epistasis detection workflows and is expected to increase replication success with biological relevance. In this contribution, we take a recently introduced regression-based epistasis detection tool as a leading example to review the key elements that need to be considered to fully appreciate the value of analytical epistasis detection performance assessments. We point out unresolved hurdles and give our perspectives towards overcoming these.

Epiregulin (EREG) and human V-ATPase (TCIRG1): genetic variation, ethnicity and pulmonary tuberculosis susceptibility in Guinea-Bissau and The Gambia.

We analyzed two West African samples (Guinea-Bissau: n=289 cases and 322 controls; The Gambia: n=240 cases and 248 controls) to evaluate single-nucleotide polymorphisms (SNPs) in Epiregulin (EREG) and V-ATPase (T-cell immune regulator 1 (TCIRG1)) using single and multilocus analyses to determine whether previously described associations with pulmonary tuberculosis (PTB) in Vietnamese and Italians would replicate in African populations. We did not detect any significant single locus or haplotype associations in either sample. We also performed exploratory pairwise interaction analyses using Visualization of Statistical Epistasis Networks (ViSEN), a novel method to detect only interactions among multiple variables, to elucidate possible interaction effects between SNPs and demographic factors. Although we found no strong evidence of marginal effects, there were several significant pairwise interactions that were identified in either the Guinea-Bissau or the Gambian samples, two of which replicated across populations. Our results indicate that the effects of EREG and TCIRG1 variants on PTB susceptibility, to the extent that they exist, are dependent on gene-gene interactions in West African populations as detected with ViSEN. In addition, epistatic effects are likely to be influenced by inter- and intra-population differences in genetic or environmental context and/or the mycobacterial lineages causing disease.

Entropy-based information gain approaches to detect and to characterize gene-gene and gene-environment interactions/correlations of complex diseases.

For complex diseases, the relationship between genotypes, environment factors, and phenotype is usually complex and nonlinear. Our understanding of the genetic architecture of diseases has considerably increased over the last years. However, both conceptually and methodologically, detecting gene-gene and gene-environment interactions remains a challenge, despite the existence of a number of efficient methods. One method that offers great promises but has not yet been widely applied to genomic data is the entropy-based approach of information theory. In this article, we first develop entropy-based test statistics to identify two-way and higher order gene-gene and gene-environment interactions. We then apply these methods to a bladder cancer data set and thereby test their power and identify strengths and weaknesses. For two-way interactions, we propose an information gain (IG) approach based on mutual information. For three-ways and higher order interactions, an interaction IG approach is used. In both cases, we develop one-dimensional test statistics to analyze sparse data. Compared to the naive chi-square test, the test statistics we develop have similar or higher power and is robust. Applying it to the bladder cancer data set allowed to investigate the complex interactions between DNA repair gene single nucleotide polymorphisms, smoking status, and bladder cancer susceptibility. Although not yet widely applied, entropy-based approaches appear as a useful tool for detecting gene-gene and gene-environment interactions. The test statistics we develop add to a growing body methodologies that will gradually shed light on the complex architecture of common diseases.

Metabolomics Insights in Early Childhood Caries.

Dental caries is characterized by a dysbiotic shift at the biofilm-tooth surface interface, yet comprehensive biochemical characterizations of the biofilm are scant. We used metabolomics to identify biochemical features of the supragingival biofilm associated with early childhood caries (ECC) prevalence and severity. The study's analytical sample comprised 289 children ages 3 to 5 (51% with ECC) who attended public preschools in North Carolina and were enrolled in a community-based cross-sectional study of early childhood oral health. Clinical examinations were conducted by calibrated examiners in community locations using International Caries Detection and Classification System (ICDAS) criteria. Supragingival plaque collected from the facial/buccal surfaces of all primary teeth in the upper-left quadrant was analyzed using ultra-performance liquid chromatography-tandem mass spectrometry. Associations between individual metabolites and 18 clinical traits (based on different ECC definitions and sets of tooth surfaces) were quantified using Brownian distance correlations (dCor) and linear regression modeling of log2-transformed values, applying a false discovery rate multiple testing correction. A tree-based pipeline optimization tool (TPOT)-machine learning process was used to identify the best-fitting ECC classification metabolite model. There were 503 named metabolites identified, including microbial, host, and exogenous biochemicals. Most significant ECC-metabolite associations were positive (i.e., upregulations/enrichments). The localized ECC case definition (ICDAS ≥1 caries experience within the surfaces from which plaque was collected) had the strongest correlation with the metabolome (dCor P = 8 × 10-3). Sixteen metabolites were significantly associated with ECC after multiple testing correction, including fucose (P = 3.0 × 10-6) and N-acetylneuraminate (p = 6.8 × 10-6) with higher ECC prevalence, as well as catechin (P = 4.7 × 10-6) and epicatechin (P = 2.9 × 10-6) with lower. Catechin, epicatechin, imidazole propionate, fucose, 9,10-DiHOME, and N-acetylneuraminate were among the top 15 metabolites in terms of ECC classification importance in the automated TPOT model. These supragingival biofilm metabolite findings provide novel insights in ECC biology and can serve as the basis for the development of measures of disease activity or risk assessment.

Electrophysiology-based stratification of pancreatic tumorigenicity by label-free single-cell impedance cytometry.

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive cancer lacking specific biomarkers that can be correlated to disease onset, promotion and progression. To assess whether tumor cell electrophysiology may serve as a marker for PDAC tumorigenicity, we use multi-frequency impedance cytometry at high throughput (∼350 cells/s) to measure the electrical phenotype of single PDAC tumor cells from xenografts, which are derived from primary pancreatic tumors versus those from liver metastases of different patients. A novel phase contrast metric based on variations in the high and low frequency impedance phase responses that is related to electrophysiology of the cell interior is found to be systematically altered as a function of tumorigenicity. PDAC cells of higher tumorigenicity exhibited lowered interior conductivity and enhanced permittivity, which is validated by the dielectrophoresis on the respective cell types. Using genetic analysis, we suggest the role of dysregulated Na+ transport and removal of Ca2+ ions from the cytoplasm on key oncogenic KRAS-driven processes that may be responsible for lowering of the interior cell conductivity. We envision that impedance cytometry can serve as a tool to quantify phenotypic heterogeneity for rapidly stratifying tumorigenicity. It can also aid in protocols for dielectrophoretic isolation of cells with a particular phenotype for prognostic studies on patient survival and to tailor therapy selection to specific patients.

Integrated analysis of genetic and proteomic data identifies biomarkers associated with adverse events following smallpox vaccination.

Complex clinical outcomes, such as adverse reaction to vaccination, arise from the concerted interactions among the myriad components of a biological system. Therefore, comprehensive etiological models can be developed only through the integrated study of multiple types of experimental data. In this study, we apply this paradigm to high-dimensional genetic and proteomic data collected to elucidate the mechanisms underlying the development of adverse events (AEs) in patients after smallpox vaccination. As vaccination was successful in all of the patients under study, the AE outcomes reported likely represent the result of interactions among immune system components that result in excessive or prolonged immune stimulation. In this study, we examined 1442 genetic variables (single nucleotide polymorphisms) and 108 proteomic variables (serum cytokine concentrations) to model AE risk. To accomplish this daunting analytical task, we employed the Random Forests (RF) method to filter the most important attributes, then we used the selected attributes to build a final decision tree model. This strategy is well suited to integrated analysis, as relevant attributes may be selected from categorical or continuous data. Importantly, RF is a natural approach for studying the type of gene-gene, gene-protein and protein-protein interactions we hypothesize to be involved in the development of clinical AEs. RF importance scores for particular attributes take interactions into account, and there may be interactions across data types. Combining information from previous studies on AEs related to smallpox vaccination with the genetic and proteomic attributes identified by RF, we built a comprehensive model of AE development that includes the cytokines intercellular adhesion molecule-1 (ICAM-1 or CD54), interleukin-10 (IL-10), and colony stimulating factor-3 (CSF-3 or G-CSF) and a genetic polymorphism in the cytokine gene interleukin-4 (IL4). The biological factors included in the model support our hypothesized mechanism for the development of AEs involving prolonged stimulation of inflammatory pathways and an imbalance of normal tissue damage repair pathways. This study shows the utility of RF for such analytical tasks, while both enhancing and reinforcing our working model of AE development after smallpox vaccination.

Aboriginal Indian residence patterns preserved in censuses and allotments.

Early reservation annuity censuses and allotment ledgers, analyzed in concert, allow identification of sociologically significant subdivisions of Native American tribes. Using this method, Southern Cheyenne manhao or "bands" can be located on the allotment map of 1892 as discrete clusters of individuals known by name, age, and sex. Measurement of linear distances among individual allotments of family members enables us to quantify jural rules of postmarital residence and confirms in a test case that the descendents of the bands at the Sand Creek Massacre in fact resided matrilocally.

Ochratoxin A: inhibition of mitochondrial respiration.

Ochratoxin A is a fungal metabolite which induces pathological changes in animals. The toxin was isolated from cultures of Aspergillus ochraceus and purified by thin-layer chromatography. Ochratoxin A and one of its hydrolysis products, dihydroisocoumarin, severely inhibited coupled respiration when applied at low concentration to rat liver mitochondria.

Neuromuscular excitability changes in the vastus medialis following anterior cruciate ligament reconstruction.

PURPOSE: Quadriceps weakness following anterior cruciate ligament reconstruction (ACLR) is prevalent despite intensive rehabilitation. Diminished neuromuscular excitability is one potential factor that may limit muscular recovery following injury or surgery. The H-reflex provides a measure of alpha motorneuron (neuromuscular) excitability in the sensory-motor pathway of the respective muscle and nerve. To date the vastus medialis (VM) and soleus (SOL) H-reflexes have been examined primarily in control subjects with induced knee joint effusion. This prospective, randomized clinical trial evaluated the affect of ACLR, utilizing hamsting (HS) or bone-patellar tendon-bone (BTB) autograft, on VM and SOL H-reflex latency and amplitude in twenty subjects. METHODS: Preoperatively bilateral VM and SOL H-reflex tests were conducted. VM and SOL H-reflexes were subsequently conducted on the involved lower extremity at 1 and 3 months post surgery. At each test session subjects completed visual analog scales and knee girth was measured. RESULTS: The VM H-reflex amplitude increased in the HS group at 3 months compared to 1-month post surgery (p<.05). Significant changes over time were also noted in the visual analog pain and functional scales and the mid-patella girth. CONCLUSIONS: The increased VM H-reflex amplitude at 3 months following HS autograft ACLR demonstrates an increase in VM neuromuscular excitability. Increased VM neuromuscular excitability was not evident in patients following BTB reconstruction. The increased neuromuscular excitability, observed only in the HS group, warrants consideration when selecting graft type for patients with extensive preoperative quadriceps dysfunction.

Male-female differences in the genetic regulation of t-PA and PAI-1 levels in a Ghanaian population.

Tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) directly influence thrombus formation and degradation, and have been identified as risk factors for thromboembolic disease. Prior studies investigated determinants of t-PA and PAI-1 expression, but mainly in Caucasian subjects. The aim of this study was to identify the contributions of genetic and other factors to inter-individual variation in plasma levels of t-PA and PAI-1 in a large-scale population-based sample from urban West Africa. t-PA, PAI-1 and several demographic, anthropometric, and metabolic parameters were measured in 992 residents of Sunyani, the capital of the Brong-Ahafo region of Ghana. In addition, nine gene polymorphisms associated with components of the renin-angiotensin and fibrinolytic systems were determined. We found that BMI, systolic and diastolic blood pressure, total cholesterol, glucose, and triglycerides were all significant predictors of t-PA and PAI-1 in both females and males. In addition, a significant relationship was found between the PAI-1 4G/5G (rs1799768) polymorphism on PAI-1 levels in females, the TPA I/D (rs4646972) polymorphism on t-PA and PAI-1 in males, the renin (rs3730103) polymorphism on t-PA and PAI-1 in males, the ethanolamine kinase 2 (rs1917542) polymorphism on PAI-1 in males, and the renin (rs1464816) polymorphism on t-PA in females and on PAI-1 in males. This study of urban West Africans shows that t-PA and PAI-1 levels are determined by both genetic loci of the fibrinolytic and renin-angiotensin systems and other factors often associated with cardiovascular disease, and that genetic factors differ between males and females.

Evaporative cooling feature selection for genotypic data involving interactions.

MOTIVATION: The development of genome-wide capabilities for genotyping has led to the practical problem of identifying the minimum subset of genetic variants relevant to the classification of a phenotype. This challenge is especially difficult in the presence of attribute interactions, noise and small sample size. METHODS: Analogous to the physical mechanism of evaporation, we introduce an evaporative cooling (EC) feature selection algorithm that seeks to obtain a subset of attributes with the optimum information temperature (i.e. the least noise). EC uses an attribute quality measure analogous to thermodynamic free energy that combines Relief-F and mutual information to evaporate (i.e. remove) noise features, leaving behind a subset of attributes that contain DNA sequence variations associated with a given phenotype. RESULTS: EC is able to identify functional sequence variations that involve interactions (epistasis) between other sequence variations that influence their association with the phenotype. This ability is demonstrated on simulated genotypic data with attribute interactions and on real genotypic data from individuals who experienced adverse events following smallpox vaccination. The EC formalism allows us to combine information entropy, energy and temperature into a single information free energy attribute quality measure that balances interaction and main effects. AVAILABILITY: Open source software, written in Java, is freely available upon request.

Gender-specific correlations of plasminogen activator inhibitor-1 and tissue plasminogen activator levels with cardiovascular disease-related traits.

BACKGROUND: The purpose of this study was to examine the correlations between plasma levels of plasminogen activator inhibitor-1 (PAI-1) and tissue plasminogen activator (t-PA) and cardiovascular disease-related traits in a general population and whether these correlations differed between females and males. METHODS: Plasma PAI-1 and t-PA antigen levels and C-reactive protein (CRP), HDL-cholesterol, triglycerides, total cholesterol, systolic blood pressure, diastolic blood pressure, urinary albumin excretion, and glucose were measured in the population-based PREVEND study in Groningen, the Netherlands (n = 2527). RESULTS: Except for CRP and total cholesterol levels, all traits were significantly different between gender (P < 0.001). PAI-1 levels were correlated with all measured cardiovascular disease-related traits (P < 0.01) in both females and males. Except for urinary albumin excretion, similar results, albeit less significant, were found for t-PA levels. Age-adjusted correlations between PAI-1 and CRP, triglycerides, total cholesterol, systolic blood pressure, and diastolic blood pressure differed significantly between females and males (P < 0.01). Many of the gender differences were predominantly present between premenopausal females and males. CONCLUSION: PAI-1 and t-PA levels were correlated with cardiovascular disease-related traits in subjects obtained from the general population and several of these correlations differed across gender. The correlations found in the present study suggest the presence of coordinated patterns of cardiovascular risk factors and indicate which traits might influence PAI-1 and t-PA levels and thereby provide a framework and potential tool for therapeutic intervention to reduce thromboembolic events in the general population.

Hybrid grammar-based approach to nonlinear dynamical system identification from biological time series.

We introduce a grammar-based hybrid approach to reverse engineering nonlinear ordinary differential equation models from observed time series. This hybrid approach combines a genetic algorithm to search the space of model architectures with a Kalman filter to estimate the model parameters. Domain-specific knowledge is used in a context-free grammar to restrict the search space for the functional form of the target model. We find that the hybrid approach outperforms a pure evolutionary algorithm method, and we observe features in the evolution of the dynamical models that correspond with the emergence of favorable model components. We apply the hybrid method to both artificially generated time series and experimentally observed protein levels from subjects who received the smallpox vaccine. From the observed data, we infer a cytokine protein interaction network for an individual's response to the smallpox vaccine.

Positional distribution of exogenous and endogenous fatty acids in triacylglycerols formed by rat adipocytes in vitro.

Rat adipocytes were used in vitro to compare the positional distributions of fatty acids of intra- and extra-cellular origin in triacyl-sn-glycerols. Fatty acids of extracellular origin were esterified to each position in similar, but not identical, proportions to the natural distributions. A high proportion of the oleic acid synthesised in the tissue by desaturation of exogenous stearic acid was found in position sn-3. When palmitic acid was the only fatty acid added, tripalmitoylglycerol was synthesised by the adipocytes. The rate and pattern of fatty acids synthesised de novo from acetate was dependent on the age of the donor rat and the concentration of acetate and presence or absence of long-chain fatty acids in the medium. The newly synthesised fatty acids were esterified in very different proportions from the natural distributions and thus from those of extracellular fatty acids. The results are discussed in terms of esterification of the fatty acids from the two sources in different compartments of the cell.

Esterification of exogenous and endogenous fatty acids by rat adipocytes in vitro.

Rat adipocytes were used in vivo to compare the esterification of exogenous fatty acids and fatty acids formed de novo from glucose or acetate. Pure single fatty acids added to the medium were esterified at comparable rates but marked differences were observed when the same acids were supplied as components of a fatty acid mixture of a composition similar to that in the tissue. Fatty acids synthesised de novo from acetate by adipocytes in a medium containing high concentrations of acetate were located predominantly in diacylglycerols. The effect was most marked with adipocytes from older rats and was enhanced by the presence of exogenous long-chain fatty acids. Exogenous oleic acid was esterified predominantly into triacylglycerols at all concentrations of acetate. No such accumulation of endogenously-synthesised fatty acids in diacylglycerols occurred when glucose was the precursor for fatty acid synthesis. The diacylglycerols formed were almost entirely of the sn-1,2-configuration.

Plasma lipids and diet of the Mvskoke Indians.

Plasma lipids, lipoproteins, and apolipoproteins were measured in 123 female and 57 male Mvskoke Indians, a population of American Indians with a high prevalence of non-insulin-dependent diabetes mellitus (NIDDM). Dietary patterns were assessed with a food-frequency questionnaire. There were no differences in total cholesterol, low-density-lipoprotein cholesterol (LDL-C), high-density-lipoprotein cholesterol (HDL-C), apolipoproteins A-I or B in female Indians with and without diabetes. In males with diabetes, however, LDL-C was lower. Triglyceride and fasting plasma glucose were higher in subjects with diabetes. Total cholesterol and LDL-C were lower and HDL-C was higher than age and sex-matched Lipid Research Clinics values, especially for subjects with diabetes. This is surprising given that the diet of Mvskoke Indians contains foods high in total fat, saturated fatty acids, and cholesterol. We may explain, in part, the low incidence of coronary heart disease in this population.

Digital ischemia in the neonate following intravenous therapy.

Ischemia and gangrene in the hand can result from attempted intravenous cannulation in the neonate. This disastrous complication can be avoided with an understanding of the blood supply of the hand. Once ischemia occurs, amputation is the usual end result, but it may be averted by early diagnosis and treatment.

Alterations of K-ras, p53, and erbB-2/neu in human lung adenocarcinomas.

The development of human adenocarcinoma of the lung involves multiple genetic changes including activation of oncogenes and loss of tumor suppressor genes. Patients whose lung tumors contain K-ras oncogene mutation, accumulation of the protein product of the tumor suppressor gene p53, or erbB-2/neu oncoprotein overexpression have been shown to have a worse prognosis. We examined these three genetic indicators in 29 lung adenocarcinomas to determine whether these markers are present in the same tumors or if they represent molecular changes that define different subsets of patients. P53 nuclear protein accumulation and erbB-2/neu protein overexpression were determined by immunohistochemical analysis of cryostat sections of tumor specimens and corresponding normal lung tissue. K-ras mutations were detected by radiolabeled oligonucleotide probes, specific for the various twelfth codon mutations, hybridized to exon 1 of K-ras, which was amplified by the polymerase chain reaction. Increased nuclear accumulation of p53 protein was found in 11 adenocarcinomas (38%). All of the p53 positive tumors were found to show high level staining and homogeneous expression of erbB-2/neu protein. K-ras mutations were detected in seven tumors (24%), all of which overexpressed erbB-2/neu. The presence of a K-ras mutation did not correlate with p53 accumulation. In total, 93% of the tumors were found to overexpress erbB-2/neu, the highest being in one tumor with erbB-2/neu gene amplification. The presence of K-ras twelfth codon mutation was associated with increased cigarette smoking. In conclusion, erbB-2/neu overexpression is a common event in lung adenocarcinomas. Furthermore, the presence of K-ras mutation and p53 protein accumulation define separate groups of patients. The mechanisms by which these genetic alterations interact or adversely affect prognosis is unknown.

Predictors of interindividual variation in ambulatory blood pressure and their time or activity dependence.

The objectives of this study were to determine whether total interindividual variation in blood pressure (BP) differs between inactive and active hours of the day, to identify predictors of interindividual variation in BP, and to assess whether variation associated with any of these identified predictors is greater (or less) during inactive hours than during active hours of the day. We obtained ambulatory BP recordings over 20 consecutive hours (12 active, out of bed [daytime]; and 8 inactive, in bed [nighttime]) in a sample of 240 unrelated, non-Hispanic white adults (138 men; 102 women). We estimated total interindividual variation in BP, and the percentage of interindividual variation associated with measures of age and body size, metabolic traits, catecholamines, erythrocyte cation transport, and renal function. We used linear regression to assess changes in the hourly estimates of total interindividual variation and in variation attributable to each set of predictor traits over the 20 h. In both men and women, total interindividual variation in systolic BP was significantly greater (not less) during inactive hours than during active hours. In addition, in women, total interindividual variation in diastolic BP was as great during inactive hours as during active hours. Each set of traits considered predicted a statistically significant percentage of interindividual variation in BP. None of the sets of traits predicted a greater percentage of interindividual variation during the inactive hours than during the active hours. Measures of age and body size, catecholamines, cation transport and renal function traits predicted significantly less interindividual variation during inactive hours than during active hours of the day. That total interindividual variation in BP is as great or greater during inactive hours than during active hours of the day emphasizes the potential for differences in nighttime BP to contribute to the development of cardiovascular disease. In as much as the predictors of interindividual variation in BP differ between the daytime and nighttime, the causes of variation during these two times may also differ.

Effect of time of day on intraindividual variability in ambulatory blood pressure.

The aim of this study was to determine whether intraindividual blood pressure (BP) variability, measured by noninvasive ambulatory monitoring, differs between the active (daytime) and inactive (nighttime) periods of the day. We obtained ambulatory BP recordings in 143 healthy adults (95 men, 48 women) from Rochester, Minnesota. Readings were obtained every 10 min for a 24-h period. We calculated the standard deviation of each individual's BP readings about the means for the active period and for the inactive period as measures of intraindividual BP variability. In men, mean within-individual standard deviations for both systolic (SBP) and diastolic blood pressure (DBP) were significantly greater during the inactive period than during the active period (for SBP: 10.3 +/- 2.1 v 11.9 +/- 2.7, P < .0001; for DBP: 8.8 +/- 2.0 v 9.7 +/- 2.5, P = .0027). In women, the mean within-individual standard deviation for SBP did not differ significantly between the active and inactive periods (9.7 +/- 2.2 v 10.3 +/- 2.4, P = 0.225) but for DBP was significantly greater during the inactive period than during the active period (8.1 +/- 2.0 v 9.2 +/- 2.3, P = .020). Statistically significant predictors of intraindividual BP variability included measures of age and body size, metabolic traits, neuroendocrine traits, erythrocyte cation traits, and renal function traits. This study demonstrates that intraindividual BP variability, as measured by noninvasive ambulatory monitoring, is as great or greater during the inactive period as during the active period of the day.