Zahra Montazeri

Systematic Meta-Analyses and Field Synopsis of Genetic Association Studies in Colorectal Cancer

BACKGROUND Colorectal cancer is a major global public health problem, with approximately 950,000 patients newly diagnosed each year. We report the first comprehensive field synopsis and creation of a parallel publicly available and regularly updated database (CRCgene) that catalogs all genetic association studies on colorectal cancer (http://www.chs.med.ed.ac.uk/CRCgene/). METHODS We performed two independent systematic reviews, reviewing 10 145 titles, then collated and extracted data from 635 publications reporting on 445 polymorphisms in 110 different genes. We carried out meta-analyses to derive summary effect estimates for 92 polymorphisms in 64 different genes. For assessing the credibility of associations, we applied the Venice criteria and the Bayesian False Discovery Probability (BFDP) test. RESULTS We consider 16 independent variants at 13 loci (MUTYH, MTHFR, SMAD7, and common variants tagging the loci 8q24, 8q23.3, 11q23.1, 14q22.2, 1q41, 20p12.3, 20q13.33, 3q26.2, 16q22.1, and 19q13.1) to have the most highly credible associations with colorectal cancer, with all variants except those in MUTYH and 19q13.1 reaching genome-wide statistical significance in at least one meta-analysis model. We identified less-credible (higher heterogeneity, lower statistical power, BFDP >0.2) associations with 23 more variants at 22 loci. The meta-analyses of a further 20 variants for which associations have previously been reported found no evidence to support these as true associations. CONCLUSION The CRCgene database provides the context for genetic association data to be interpreted appropriately and helps inform future research direction.

Waterpipe smoking and cancer: systematic review and meta-analysis.

Objective Although accumulating evidence suggests harmful effects of waterpipe smoking, there is limited information about its direct association with chronic diseases, notably cancer. We provide an up-to-date systematic review and meta-analysis of the association between waterpipe smoking and cancer. Data sources Systematic search of articles indexed in main biomedical databases: Pubmed, EmBase, Google Scholar and Web of Science, published between 1962 and September 2014. Search keywords included a combination of waterpipe or hookah, sheesha, nargile, hubble-bubble, goza or gaylan, and cancer. Study selection Focus on observational studies (cohort, case–control, cross-sectional) that evaluated the association between waterpipe smoking and cancer. Studies with mixed exposures excluded. Data extraction Two investigators independently extracted data and reached consensus on all items. Data synthesis 13 case–control studies met the inclusion criteria and were considered for meta-analysis. The methodological quality of included studies was assessed using the Newcastle-Ottawa Scale (NOS). Meta-analysis revealed a positive association between waterpipe smoking and lung cancer (OR=4.58 (2.61 to 8.03); I2=44.67%), and oesophageal cancer (OR=3.63 (1.39 to 9.44); I2 =94.49%). The majority of studies had a NOS score of 5–6 or 7, indicating ‘fair’ or ‘good’ quality, respectively. Conclusions Our findings support a positive association between waterpipe smoking and cancer risk. However, high-quality studies with standardised exposure measurements are needed to clarify the contribution of waterpipe smoking to chronic diseases. More investments in initiatives for surveillance, intervention and regulatory policy for waterpipe smoking are urgently warranted.

Shrinkage Estimation of Effect Sizes as an Alternative to Hypothesis Testing Followed by Estimation in High-Dimensional Biology: Applications to Differential Gene Expression

Research on analyzing microarray data has focused on the problem of identifying differentially expressed genes to the neglect of the problem of how to integrate evidence that a gene is differentially expressed with information on the extent of its differential expression. Consequently, researchers currently prioritize genes for further study either on the basis of volcano plots or, more commonly, according to simple estimates of the fold change after filtering the genes with an arbitrary statistical significance threshold. While the subjective and informal nature of the former practice precludes quantification of its reliability, the latter practice is equivalent to using a hard-threshold estimator of the expression ratio that is not known to perform well in terms of mean-squared error, the sum of estimator variance and squared estimator bias. On the basis of two distinct simulation studies and data from different microarray studies, we systematically compared the performance of several estimators representing both current practice and shrinkage. We find that the threshold-based estimators usually perform worse than the maximum-likelihood estimator (MLE) and they often perform far worse as quantified by estimated mean-squared risk. By contrast, the shrinkage estimators tend to perform as well as or better than the MLE and never much worse than the MLE, as expected from what is known about shrinkage. However, a Bayesian measure of performance based on the prior information that few genes are differentially expressed indicates that hard-threshold estimators perform about as well as the local false discovery rate (FDR), the best of the shrinkage estimators studied. Based on the ability of the latter to leverage information across genes, we conclude that the use of the local-FDR estimator of the fold change instead of informal or threshold-based combinations of statistical tests and non-shrinkage estimators can be expected to substantially improve the reliability of gene prioritization at very little risk of doing so less reliably. Since the proposed replacement of post-selection estimates with shrunken estimates applies as well to other types of high-dimensional data, it could also improve the analysis of SNP data from genome-wide association studies.

Deamination intensity profiling of human APOBEC3 protein activity along the near full-length genomes of HIV-1 and MoMLV by HyperHRM analysis.

Enzymatic deamination of cytidines in DNA is an intrinsic component of antibody maturation and retroviral resistance, but can also be a source of HIV drug resistance and cancer-causing mutations. Here, we developed a high-throughput method based on high resolution melt (HRM) analysis called HyperHRM that can screen genomic DNA for rare hypermutated proviral sequences and accurately quantify the number of C-to-T or G-to-A mutations in each sequence. We demonstrate the effectiveness of the approach by profiling in parallel the intensity of the DNA mutator activity of all seven human APOBEC3 proteins on the near full-length sequence of HIV-1 and the Moloney murine leukemia virus. Additionally, HRM was successfully used to identify hypermutated proviral sequences in peripheral blood mononuclear cells from an HIV-1 patient. These results exemplify the effectiveness of HRM-based approaches for hypermutation quantification and for the detection of hypermutated DNA sequences potentially associated with disease or retroviral drug resistance.

Gene network reconstruction from transcriptional dynamics under kinetic model uncertainty

Motivation: Measurements of gene expression over time enable the reconstruction of transcriptional networks. However, Bayesian networks and many other current reconstruction methods rely on assumptions that conflict with the differential equations that describe transcriptional kinetics. Practical approximations of kinetic models would enable inferring causal relationships between genes from expression data of microarray, tag-based and conventional platforms, but conclusions are sensitive to the assumptions made. Results: The representation of a sufficiently large portion of genome enables computation of an upper bound on how much confidence one may place in influences between genes on the basis of expression data. Information about which genes encode transcription factors is not necessary but may be incorporated if available. The methodology is generalized to cover cases in which expression measurements are missing for many of the genes that might control the transcription of the genes of interest. The assumption that the gene expression level is roughly proportional to the rate of translation led to better empirical performance than did either the assumption that the gene expression level is roughly proportional to the protein level or the Bayesian model average of both assumptions. Availability: http://www.oisb.ca points to R code implementing the methods (R Development Core Team 2004). Contact: dbickel@uottawa.ca Supplementary information: http://www.davidbickel.com

Family history: impact on coronary heart disease risk assessment beyond guideline-defined factors.

Objective: Family history (FH) provides insights into the effects of shared genomic susceptibilities, environments and behaviors, making it a potentially valuable risk assessment tool for chronic diseases. We assessed whether coronary heart disease (CHD) risk assessment is improved when FH information is added to other clinical information recommended in guidelines. Methods: We applied logistic regression analyses to cross-sectional data originally obtained from a UK study of women who delivered a live-born infant between 1951 and 1970. We developed 3 models: Model 1 included only the covariates in a guideline applicable to the population, Model 2 added FH to Model 1, and Model 3 included a fuller range of risk factors. For each model, its ability to discriminate between study subjects with and those without CHD was evaluated and its impact on risk classification examined using the net reclassification index. Results: FH was an independent risk factor for CHD (odds ratio = 1.7, 95% confidence interval = 1.26-2.47) and improved discrimination beyond guideline-defined clinical factors (p < 0.0006). However, the difference in the area under the curve of 2.8% and the extent of patient reclassification resulting from the inclusion of FH were small (p = 0.11). Conclusion: While FH were a significant independent risk factor for CHD, it added little to risk factors typically included in guidelines.

Empirical measures for incomplete data with applications

Methods are proposed to construct empirical measures when there are missing terms among the components of a random vector. Fur- thermore, Vapnik-Chevonenkis type exponential bounds are obtained on the uniform deviations of these estimators, from the true probabilities. These results can then be used to deal with classical problems such as statistical classification, via empirical risk minimization, when there are missing covariates among the data. Another application involves the uni- form estimation of a distribution function. AMS 2000 subject classifications: Primary 60G50, 62G15; secondary 62H30.

Aggregating classifiers via Rademacher–Walsh polynomials

Methods are proposed to combine several individual classifiers in order to develop more accurate classification rules. The proposed algorithm uses Rademacher–Walsh polynomials to combine M (≥2) individual classifiers in a nonlinear way. The resulting classifier is optimal in the sense that its misclassification error rate is always less than, or equal to, that of each constituent classifier. A number of numerical examples (based on both real and simulated data) are also given. These examples demonstrate some new, and far-reaching, benefits of working with combined classifiers.

Effect of childhood BMI on asthma: a systematic review and meta-analysis of case-control studies

BackgroundAsthma is a multifactorial syndrome that threatens the health of children. Body mass index (BMI) might be one of the potential factors but the evidence is controversial. The aim of this study is to perform a comprehensive meta-analysis to investigate the association between asthma and BMI.MethodsElectronic databases including, Web of Science, Pubmed, Scopus, Science Direct, ProQuest, up to April 2017, were searched by two researchers independently. The keywords “asthma, body mass index, obesity, overweight, childhood and adolescence” were used. Random and fixed effects models were applied to obtain the overall odds ratios (ORs) and standardized mean difference (SMD). Heterogeneity between the studies was examined using I2 and Cochrane Q statistics.ResultsAfter reviewing 2511 articles, 16 studies were eligible for meta-analysis according to inclusion/exclusion criteria. A meta-analysis from 11 case-control studies revealed OR of asthma and overweight as OR = 1.64; (95% Confidence Interval (CI): 1.13–2.38) and from 14 case-control studies, OR for asthma and obesity was OR = 1.92 (95% CI: 1.39–2.65), which indicated that risk of asthma in overweight and obese children and adolescence was significantly higher (1.64 and 1.92 times) than that of individuals with (p-value < 0.01 for underweight/normal weight in both cases). Furthermore, there was a significant relationship between asthma and BMI > 85 percentile according to SMD SMD = 0.21; (95%CI: 0.03–0.38; p-value = 0.021).ConclusionsThe results showed a significant relationship between BMI (obesity/overweight) and asthma among children and adolescents. It is important to study the confounding factors that affect the relationship between asthma and BMI in future epidemiological researches.

Prevalence of low back pain in Iranian nurses: a systematic review and meta-analysis

BackgroundLow back pain (LBP) as a musculoskeletal disorder is one of the most common occupational injuries in nurses but there isn’t any valid measure of the prevalence of LBP in Iranian nursing. In order to increase the power and improve the estimates of the prevalence of LBP in Iranian nurses, a comprehensive meta-analysis was carried out. A summary measure of all studies conducted in this field was found and distributions of LBP were evaluated based on different variables.MethodsInclusion criteria included articles with prevalence of LBP in Iranian nurses, who had at least six months of work experience without any trauma, injuries to spine, or any underlying disease. The keywords“prevalence, low back pain, nurses”, and “Iran” were used as part of this search. Databases such as Pubmed, Web of Science, Science direct, Scopus, IranMedex, Irandoc, Magiran, SID, CIVILICA, IMEMR and Google scholar were searched up to and including 15 June 2016. For data extraction a form was designed that included the following variables: Author names, province, sample size, age, gender, marital status, work experience, body mass index, job type, smoking status, work schedule, year of publication, type of standard questionnaire, prevalence of LBP, studies’ quality score and climate classifications. Data analysis was carried out using fixed and random effects model. Heterogeneity between studies was assessed by using the I2 and Q tests.ResultsIn all 1250 articles were identified and 22 articles with 9347 participants met the inclusion criteria for meta-analyses after filtering. The prevalence of low back pain during their working life and during the last year, was estimated at 63% (95% Confidence Interval (CI): 57.4–68.5) and 61.2% (95% CI: 55.7–66.7) respectively. The prevalence rate of this disorder was 58.7% (95% CI: 35.8–81.7) and 60.4% (95% CI: 52.2–68.6) among men and women respectively. Furthermore, prevalence’s of LBP were 59.5% in wards nurses, 50.3% in operating room technicians, and 39.4% in aid nurses.ConclusionsThe results showed the high prevalence of LBP injury in nurses, especially female nurses. The effect of musculoskeletal disorders such as LBP may be reduced by considering proper observation of the principles of ergonomics in the workplace, performing physical examinations on a regular basis, identifying risk factors in the advancement of musculoskeletal disorders and then trying to fix them.