Statistical Impact of Sample Size and Imbalance on Multivariate Analysis in silico and A Case Study in the UK Biobank.
Ontology highlight
ABSTRACT:
SUBMITTER: Zhang X
PROVIDER: S-EPMC8075427 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BACKGROUND:The development of sequencing techniques and statistical methods provides great opportunities for identifying the impact of rare genetic variation on complex traits. However, there is a lack of knowledge on the impact of sample size, case numbers, the balance of cases vs controls for both burden and dispersion based rare variant association methods. For example, Phenome-Wide Association Studies may have a wide range of case and control sample sizes across hundreds of diagnoses and traits, and with the application of statistical methods to rare variants, it is important to understand the strengths and limitations of the analyses. RESULTS:We conducted a large-scale simulation of randomly selected low-frequency protein-coding regions using twelve different balanced samples with an equal number of cases and controls as well as twenty-one unbalanced sample scenarios. We further explored statistical performance of different minor allele frequency thresholds and a range of genetic effect sizes. Our simulation results demonstrate that using an unbalanced study design has an overall higher type I error rate for both burden and dispersion tests compared with a balanced study design. Regression has an overall higher type I error with balanced cases and controls, while SKAT has higher type I error for unbalanced case-control scenarios. We also found that both type I error and power were driven by the number of cases in addition to the case to control ratio under large control group scenarios. Based on our power simulations, we observed that a SKAT analysis with case numbers larger than 200 for unbalanced case-control models yielded over 90% power with relatively well controlled type I error. To achieve similar power in regression, over 500 cases are needed. Moreover, SKAT showed higher power to detect associations in unbalanced case-control scenarios than regression. CONCLUSIONS:Our results provide important insights into rare variant association study designs by providing a landscape of type I error and statistical power for a wide range of sample sizes. These results can serve as a benchmark for making decisions about study design for rare variant analyses.
| S-EPMC6343276 | biostudies-other
Project description:In high-dimensional regression problems, often a relatively small subset of the features are relevant for predicting the outcome, and methods that impose sparsity on the solution are popular. When multiple correlated outcomes are available (multitask), reduced rank regression is an effective way to borrow strength and capture latent structures that underlie the data. Our proposal is motivated by the UK Biobank population-based cohort study, where we are faced with large-scale, ultrahigh-dimensional features, and have access to a large number of outcomes (phenotypes)-lifestyle measures, biomarkers, and disease outcomes. We are hence led to fit sparse reduced-rank regression models, using computational strategies that allow us to scale to problems of this size. We use a scheme that alternates between solving the sparse regression problem and solving the reduced rank decomposition. For the sparse regression component we propose a scalable iterative algorithm based on adaptive screening that leverages the sparsity assumption and enables us to focus on solving much smaller subproblems. The full solution is reconstructed and tested via an optimality condition to make sure it is a valid solution for the original problem. We further extend the method to cope with practical issues, such as the inclusion of confounding variables and imputation of missing values among the phenotypes. Experiments on both synthetic data and the UK Biobank data demonstrate the effectiveness of the method and the algorithm. We present multiSnpnet package, available at http://github.com/junyangq/multiSnpnet that works on top of PLINK2 files, which we anticipate to be a valuable tool for generating polygenic risk scores from human genetic studies.
| S-EPMC9454085 | biostudies-literature
Project description:Background and Objectives:Researchers typically use Cohen's guidelines of Pearson's r = .10, .30, and .50, and Cohen's d = 0.20, 0.50, and 0.80 to interpret observed effect sizes as small, medium, or large, respectively. However, these guidelines were not based on quantitative estimates and are only recommended if field-specific estimates are unknown. This study investigated the distribution of effect sizes in both individual differences research and group differences research in gerontology to provide estimates of effect sizes in the field. Research Design and Methods:Effect sizes (Pearson's r, Cohen's d, and Hedges' g) were extracted from meta-analyses published in 10 top-ranked gerontology journals. The 25th, 50th, and 75th percentile ranks were calculated for Pearson's r (individual differences) and Cohen's d or Hedges' g (group differences) values as indicators of small, medium, and large effects. A priori power analyses were conducted for sample size calculations given the observed effect size estimates. Results:Effect sizes of Pearson's r = .12, .20, and .32 for individual differences research and Hedges' g = 0.16, 0.38, and 0.76 for group differences research were interpreted as small, medium, and large effects in gerontology. Discussion and Implications:Cohen's guidelines appear to overestimate effect sizes in gerontology. Researchers are encouraged to use Pearson's r = .10, .20, and .30, and Cohen's d or Hedges' g = 0.15, 0.40, and 0.75 to interpret small, medium, and large effects in gerontology, and recruit larger samples.
| S-EPMC6736231 | biostudies-literature
Project description:Inference on disease dynamics is typically performed using case reporting time series of symptomatic disease. The inferred dynamics will vary depending on the reporting patterns and surveillance system for the disease in question, and the inference will miss mild or underreported epidemics. To eliminate the variation introduced by differing reporting patterns and to capture asymptomatic or subclinical infection, inferential methods can be applied to serological data sets instead of case reporting data. To reconstruct complete disease dynamics, one would need to collect a serological time series. In the statistical analysis presented here, we consider a particular kind of serological time series with repeated, periodic collections of population-representative serum. We refer to this study design as a serial seroepidemiology (SSE) design, and we base the analysis on our epidemiological knowledge of influenza. We consider a study duration of three to four years, during which a single antigenic type of influenza would be circulating, and we evaluate our ability to reconstruct disease dynamics based on serological data alone. We show that the processes of reinfection, antibody generation, and antibody waning confound each other and are not always statistically identifiable, especially when dynamics resemble a non-oscillating endemic equilibrium behavior. We introduce some constraints to partially resolve this confounding, and we show that transmission rates and basic reproduction numbers can be accurately estimated in SSE study designs. Seasonal forcing is more difficult to identify as serology-based studies only detect oscillations in antibody titers of recovered individuals, and these oscillations are typically weaker than those observed for infected individuals. To accurately estimate the magnitude and timing of seasonal forcing, serum samples should be collected every two months and 200 or more samples should be included in each collection; this sample size estimate is sensitive to the antibody waning rate and the assumed level of seasonal forcing.
| S-EPMC4558460 | biostudies-literature
Project description:Inadequate statistical power to detect treatment effects in health research is a problem that is compounded when testing for mediation. In general, the preferred strategy for increasing power is to increase the sample size, but there are many situations where additional participants cannot be recruited, necessitating the use of other methods to increase statistical power. Many of these other strategies, commonly applied to analysis of variance and multiple regression models, can be applied to mediation models with similar results. Additional predictors or blocking variables will increase or decrease statistical power, however, depending on whether these variables are related to the mediator, the outcome, or both. The effect of these two methods on the power for tests of mediation is illustrated through the use of simulations. Implications for health researchers using these methods are discussed.
| S-EPMC4836861 | biostudies-literature
Project description:Researchers in the social and behavioral sciences often have clear expectations about the order/direction of the parameters in their statistical model. For example, a researcher might expect that regression coefficient ?1 is larger than ?2 and ?3. The corresponding hypothesis is H: ?1 > {?2, ?3} and this is known as an (order) constrained hypothesis. A major advantage of testing such a hypothesis is that power can be gained and inherently a smaller sample size is needed. This article discusses this gain in sample size reduction, when an increasing number of constraints is included into the hypothesis. The main goal is to present sample-size tables for constrained hypotheses. A sample-size table contains the necessary sample-size at a pre-specified power (say, 0.80) for an increasing number of constraints. To obtain sample-size tables, two Monte Carlo simulations were performed, one for ANOVA and one for multiple regression. Three results are salient. First, in an ANOVA the needed sample-size decreases with 30-50% when complete ordering of the parameters is taken into account. Second, small deviations from the imposed order have only a minor impact on the power. Third, at the maximum number of constraints, the linear regression results are comparable with the ANOVA results. However, in the case of fewer constraints, ordering the parameters (e.g., ?1 > ?2) results in a higher power than assigning a positive or a negative sign to the parameters (e.g., ?1 > 0).
| S-EPMC4292225 | biostudies-literature
Project description:ObjectiveTo assess the association between household size and risk of non-severe or severe COVID-19.DesignA longitudinal observational study.SettingThis study utilised UK Biobank linked to national SARS-CoV-2 laboratory test data.Participants401,910 individuals with available data on household size in UK Biobank.Main outcome measuresHousehold size was categorised as single occupancy, two-person households and households of three or more. Severe COVID-19 was defined as a positive SARS-CoV-2 test on hospital admission or death with COVID-19 recorded as the underlying cause; and non-severe COVID-19 as a positive test from a community setting. Logistic regression models were fitted to assess associations, adjusting for potential confounders.ResultsOf 401,910 individuals, 3612 (1%) were identified as having suffered from a severe COVID-19 infection and 11,264 (2.8%) from a non-severe infection, between 16 March 2020 and 16 March 2021. Overall, the odds of severe COVID-19 was significantly higher among individuals living alone (adjusted odds ratio: 1.24 [95% confidence interval: 1.14 to 1.36], or living in a household of three or more individuals (adjusted odds ratio: 1.28 [1.17 to 1.39], when compared to individuals living in a household of two. For non-severe COVID-19 infection, individuals living in a single-occupancy household had lower odds compared to those living in a household of two (adjusted odds ratio: 0.88 [0.82 to 0.93].ConclusionsOdds of severe or non-severe COVID-19 infection were associated with household size. Increasing understanding of why certain households are more at risk is important for limiting spread of the infection.
| S-EPMC8972956 | biostudies-literature
Project description:Large studies use genotype data to discover genetic contributions to complex traits and infer relationships between those traits. Co-incident geographical variation in genotypes and health traits can bias these analyses. Here we show that single genetic variants and genetic scores composed of multiple variants are associated with birth location within UK Biobank and that geographic structure in genotype data cannot be accounted for using routine adjustment for study centre and principal components derived from genotype data. We find that major health outcomes appear geographically structured and that coincident structure in health outcomes and genotype data can yield biased associations. Understanding and accounting for this phenomenon will be important when making inference from genotype data in large studies.
| S-EPMC6338768 | biostudies-literature
Project description:Sample size calculations are an essential component of the design and evaluation of scientific studies. However, there is a lack of clear guidance for determining the sample size needed for phylogenetic studies, which are becoming an essential part of studying pathogen transmission. We introduce a statistical framework for determining the number of true infector-infectee transmission pairs identified by a phylogenetic study, given the size and population coverage of that study. We then show how characteristics of the criteria used to determine linkage and aspects of the study design can influence our ability to correctly identify transmission links, in sometimes counterintuitive ways. We test the overall approach using outbreak simulations and provide guidance for calculating the sensitivity and specificity of the linkage criteria, the key inputs to our approach. The framework is freely available as the R package phylosamp, and is broadly applicable to designing and evaluating a wide array of pathogen phylogenetic studies.
| S-EPMC8284614 | biostudies-literature
Project description:We conducted a two-sample Mendelian randomization study to explore the associations of iron status with overall cancer and 22 site-specific cancers. Single-nucleotide polymorphisms for iron status were obtained from a genome-wide association study of 48,972 European-descent individuals. Summary-level data for breast and other cancers were obtained from the Breast Cancer Association Consortium and UK Biobank. Genetically predicted iron status was positively associated with liver cancer and inversely associated with brain cancer but not associated with overall cancer or the other 20 studied cancer sites at p < 0.05. The odds ratios of liver cancer were 2.45 (95% CI, 0.81, 7.45; p = 0.11), 2.11 (1.16, 3.83; p = 0.02), 10.89 (2.44, 48.59; p = 0.002) and 0.30 (0.17, 0.53; p = 2 × 10-5) for one standard deviation increment of serum iron, transferrin saturation, ferritin and transferrin levels, respectively. For brain cancer, the corresponding odds ratios were 0.69 (0.48, 1.00; p = 0.05), 0.75 (0.59, 0.97; p = 0.03), 0.41 (0.20, 0.88; p = 0.02) and 1.49 (1.04, 2.14; p = 0.03). Genetically high iron status was positively associated with liver cancer and inversely associated with brain cancer.
| S-EPMC7071358 | biostudies-literature