Combining dependent P-values with an empirical adaptation of Brown's method.
Ontology highlight
ABSTRACT: Motivation
Combining P-values from multiple statistical tests is a common exercise in bioinformatics. However, this procedure is non-trivial for dependent P-values. Here, we discuss an empirical adaptation of Brown's method (an extension of Fisher's method) for combining dependent P-values which is appropriate for the large and correlated datasets found in high-throughput biology.Results
We show that the Empirical Brown's method (EBM) outperforms Fisher's method as well as alternative approaches for combining dependent P-values using both noisy simulated data and gene expression data from The Cancer Genome Atlas.Availability and implementation
The Empirical Brown's method is available in Python, R, and MATLAB and can be obtained from https://github.com/IlyaLab/CombiningDependentPvalues UsingEBM The R code is also available as a Bioconductor package from https://www.bioconductor.org/packages/devel/bioc/html/EmpiricalBrownsMethod.htmlContact
Theo.Knijnenburg@systemsbiology.orgSupplementary information
Supplementary data are available at Bioinformatics online.
SUBMITTER: Poole W
PROVIDER: S-EPMC5013915 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:We propose an adaptive truncated product method that facilitates the selection of the truncation point among a set of candidates. To efficiently estimate the distribution of the proposed method when the p-values are correlated, we develop a single-layer bootstrap procedure.
| S-EPMC3736994 | biostudies-literature
Project description:Rapid developments in molecular technology have yielded a large amount of high throughput genetic data to understand the mechanism for complex traits. The increase of genetic variants requires hundreds and thousands of statistical tests to be performed simultaneously in analysis, which poses a challenge to control the overall Type I error rate. Combining p-values from multiple hypothesis testing has shown promise for aggregating effects in high-dimensional genetic data analysis. Several p-value combining methods have been developed and applied to genetic data; see Dai et al. (2012b) for a comprehensive review. However, there is a lack of investigations conducted for dependent genetic data, especially for weighted p-value combining methods. Single nucleotide polymorphisms (SNPs) are often correlated due to linkage disequilibrium (LD). Other genetic data, including variants from next generation sequencing, gene expression levels measured by microarray, protein and DNA methylation data, etc. also contain complex correlation structures. Ignoring correlation structures among genetic variants may lead to severe inflation of Type I error rates for omnibus testing of p-values. In this work, we propose modifications to the Lancaster procedure by taking the correlation structure among p-values into account. The weight function in the Lancaster procedure allows meaningful biological information to be incorporated into the statistical analysis, which can increase the power of the statistical testing and/or remove the bias in the process. Extensive empirical assessments demonstrate that the modified Lancaster procedure largely reduces the Type I error rates due to correlation among p-values, and retains considerable power to detect signals among p-values. We applied our method to reassess published renal transplant data, and identified a novel association between B cell pathways and allograft tolerance.
| S-EPMC3929847 | biostudies-literature
Project description:BackgroundProcedures for priority setting need to incorporate both scientific evidence and public values. The aim of this study was to test out a model for priority setting which incorporates both scientific evidence and public values, and to explore use of evidence by a selection of stakeholders and to study reasons for the relative ranking of health care interventions in a setting of extreme resource scarcity.MethodsSystematic search for and assessment of relevant evidence for priority setting in a low-income country. Development of a balance sheet according to Eddy's explicit method. Eight group interviews (n-85), using a modified nominal group technique for eliciting individual and group rankings of a given set of health interventions.ResultsThe study procedure made it possible to compare the groups' ranking before and after all the evidence was provided to participants. A rank deviation is significant if the rank order of the same intervention differed by two or more points on the ordinal scale. A comparison between the initial rank and the final rank (before deliberation) showed a rank deviation of 67%. The difference between the initial rank and the final rank after discussion and voting gave a rank deviation of 78%.ConclusionEvidence-based and deliberative decision-making does change priorities significantly in an experimental setting. Our use of the balance sheet method was meant as a demonstration project, but could if properly developed be feasible for health planners, experts and health workers, although more work is needed before it can be used for laypersons.
| S-EPMC2096625 | biostudies-literature
Project description:In recent environmental research, relational values (RVs) have emerged as a new group of values to explain environmental behavior. Although this new concept is attracting attention, empirical studies on the subject are still rare. On this basis, we have conducted three studies to analyze an existing measurement tool for RVs and compared the construct with the concept of connection to nature. In study 1, we confirmed convergent and discriminant validity of the RV scale by comparing it with the Two Factor Model of Environmental Values (2-MEV) model using a sample of n = 350 university students. Additionally, study 1 verified reliability using test-retest reliability on three different groups of students (n 1 = 53; n 2 = 37; n 3 = 48). In study 2, principal component analyses were performed to examine the structure of RVs and to compare it to the concept of connection to nature by reusing the sample 350 university students from study 1. The results show that RVs and connection to nature are not fundamentally distinct constructs, but overlap. However, if the structure of the RV measurement is forced to a single factor, no perfect fit is found, making a multidimensional solution more likely. A third study was conducted to review the results from study 2 using confirmatory factor analysis on a new sample of 878 university and high school students. Study 3 confirmed RVs as a multidimensional construct with three factors: care, community, and connection. It also proved the overlap of the connection to nature and RV concepts to some extent.
| S-EPMC7083099 | biostudies-literature
Project description: Not available
| S-EPMC5108459 | biostudies-literature
Project description:Controversy over non-reproducible published research reporting a statistically significant result has produced substantial discussion in the literature. p-value calibration is a recently proposed procedure for adjusting p-values to account for both random and systematic errors that address one aspect of this problem. The method's validity rests on the key assumption that bias in an effect estimate is drawn from a normal distribution whose mean and variance can be correctly estimated. We investigated the method's control of type I and type II error rates using simulated and real-world data. Under mild violations of underlying assumptions, control of the type I error rate can be conservative, while under more extreme departures, it can be anti-conservative. The extent to which the assumption is violated in real-world data analyses is unknown. Barriers to testing the plausibility of the assumption using historical data are discussed. Our studies of the type II error rate using simulated and real-world electronic health care data demonstrated that calibrating p-values can substantially increase the type II error rate. The use of calibrated p-values may reduce the number of false-positive results, but there will be a commensurate drop in the ability to detect a true safety or efficacy signal. While p-value calibration can sometimes offer advantages in controlling the type I error rate, its adoption for routine use in studies of real-world health care datasets is premature. Separate characterizations of random and systematic errors provide a richer context for evaluating uncertainty surrounding effect estimates. Copyright © 2016 John Wiley & Sons, Ltd.
| S-EPMC5012943 | biostudies-literature
Project description:The derivation of protective values for aquatic life can be enhanced by the development and use of bioavailability models. Recent advances to metals bioavailability modeling are applicable to other analyte groups and should be widely considered. We conducted a meta-analysis of the available aquatic toxicity literature for fluoride to evaluate the utility of hardness, alkalinity, and chloride as toxicity-modifying factors (TMFs) in empirical bioavailability models of freshwater taxa. The resulting optimal multiple linear regression model predicting acute fluoride toxicity to the invertebrate Hyalella azteca included all three TMFs (observed vs. predicted 50% lethal concentrations, R2 = 0.88) and the optimal model predicting toxicity to the fish Oncorhynchus mykiss included alkalinity and hardness (R2 = 0.37). At >20 mg/L chloride, the preliminary final acute values for fluoride were within 1 order of magnitude and ranged from approximately 18.1 to 56.3 mg/L, depending on water chemistry. Sensitivity of H. azteca to low-chloride conditions increased model uncertainty when chloride was <20 mg/L. Because of limited toxicity data, chronic bioavailability models were not developed, and final chronic values were derived using an acute-to-chronic ratio (ACR) approach. Accounting for TMFs, the geometric mean ACR was 5.4 for fish and invertebrate taxa (n = 6). The present assessment highlights the need to expand bioavailability modeling to include inorganic anions, particularly fluoride, and demonstrates that existing promulgated protective values for fluoride are likely overly conservative. More toxicological studies are recommended to further refine multivariate empirical bioavailability models for inorganic anions. Environ Toxicol Chem 2022;41:396-409. © 2021 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.
| S-EPMC9303462 | biostudies-literature
Project description:Most libraries for fragment-based drug discovery are restricted to 1,000-10,000 compounds, but over 500,000 fragments are commercially available and potentially accessible by virtual screening. Whether this larger set would increase chemotype coverage, and whether a computational screen can pragmatically prioritize them, is debated. To investigate this question, a 1281-fragment library was screened by nuclear magnetic resonance (NMR) against AmpC ?-lactamase, and hits were confirmed by surface plasmon resonance (SPR). Nine hits with novel chemotypes were confirmed biochemically with KI values from 0.2 to low mM. We also computationally docked 290,000 purchasable fragments with chemotypes unrepresented in the empirical library, finding 10 that had KI values from 0.03 to low mM. Though less novel than those discovered by NMR, the docking-derived fragments filled chemotype holes from the empirical library. Crystal structures of nine of the fragments in complex with AmpC ?-lactamase revealed new binding sites and explained the relatively high affinity of the docking-derived fragments. The existence of chemotype holes is likely a general feature of fragment libraries, as calculation suggests that to represent the fragment substructures of even known biogenic molecules would demand a library of minimally over 32,000 fragments. Combining computational and empirical fragment screens enables the discovery of unexpected chemotypes, here by the NMR screen, while capturing chemotypes missing from the empirical library and tailored to the target, with little extra cost in resources.
| S-EPMC4215856 | biostudies-literature
Project description:Current practice of validating predicted protein structural model is knowledge-based where scoring parameters are derived from already known structures to obtain decision on validation out of this structure information. For example, the scoring parameter, Ramachandran Score gives percentage conformity with steric-property higher value of which implies higher acceptability. On the other hand, Force-Field Energy Score gives conformity with energy-wise stability higher value of which implies lower acceptability. Naturally, setting these two scoring parameters as target objectives sometimes yields a set of multiple models for the same protein for which acceptance based on a particular parameter, say, Ramachandran score, may not satisfy well with the acceptance of the same model based on other parameter, say, energy score. The confusion set of such models can further be resolved by introducing some parameters value of which are easily obtainable through experiment on the same protein. In this piece of work it was found that the confusion regarding final acceptance of a model out of multiple models of the same protein can be removed using a parameter Surface Rough Index which can be obtained through semi-empirical method from the ordinary microscopic image of heat denatured protein.
| S-EPMC3524942 | biostudies-literature
Project description:The outgroup method is widely used to root phylogenetic trees. An accurate root indication, however, strongly depends on the availability of a proper outgroup. An alternate rooting method is the midpoint rooting (MPR). In this case, the root is set at the midpoint between the two most divergent operational taxonomic units. Although the midpoint rooting algorithm has been extensively used, the efficiency of this method in retrieving the correct root remains untested. In the present study, we empirically tested the success rate of the MPR in obtaining the outgroup root for a given phylogenetic tree. This was carried out by eliminating outgroups in 50 selected data sets from 33 papers and rooting the trees with the midpoint method. We were thus able to compare the root position retrieved by each method. Data sets were separated into three categories with different root consistencies: data sets with a single outgroup taxon (54% success rate for MPR), data sets with multiple outgroup taxa that showed inconsistency in root position (82% success rate), and data sets with multiple outgroup taxa in which root position was consistent (94% success rate). Interestingly, the more consistent the outgroup root is, the more successful MPR appears to be. This is a strong indication that the MPR method is valuable, particularly for cases where a proper outgroup is unavailable.
| S-EPMC7110036 | biostudies-literature