Linemodels: clustering effects based on linear relationships.
Ontology highlight
ABSTRACT: Summary
Estimation of effects of multiple explanatory variables on multiple outcome measures has become routine across life sciences with high-throughput molecular technologies. The linemodels R-package allows a probabilistic clustering of variables based on their observed effect sizes on two outcomes.Availability and implementation
An open source implementation in R available at github.com/mjpirinen/linemodels.
SUBMITTER: Pirinen M
PROVIDER: S-EPMC10005595 | biostudies-literature | 2023 Mar
REPOSITORIES: biostudies-literature
Publications
linemodels: clustering effects based on linear relationships.
Bioinformatics (Oxford, England) 20230301 3
<h4>Summary</h4>Estimation of effects of multiple explanatory variables on multiple outcome measures has become routine across life sciences with high-throughput molecular technologies. The linemodels R-package allows a probabilistic clustering of variables based on their observed effect sizes on two outcomes.<h4>Availability and implementation</h4>An open source implementation in R available at github.com/mjpirinen/linemodels. ...[more]
Similar Datasets
Project description:Emerging evidence suggests that a genetic variant can affect multiple phenotypes, especially in complex human diseases. Therefore, joint analysis of multiple phenotypes may offer new insights into disease etiology. Recently, many statistical methods have been developed for joint analysis of multiple phenotypes, including the clustering linear combination (CLC) method. Due to the unknown number of clusters for a given data, a simulation procedure must be used to evaluate the p-value of the final test statistic of CLC. This makes the CLC method computationally demanding. In this paper, we use a stopping criterion to determine the number of clusters in the CLC method. We have named our method, hierarchical clustering CLC (HCLC). HCLC has an asymptotic distribution, which is very computationally efficient and makes it applicable for genome-wide association studies. Extensive simulations together with the COPDGene data analysis have been used to assess the type I error rates and power of our proposed method. Our simulation results demonstrate that the type I error rates of HCLC are effectively controlled in different realistic settings. HCLC either outperforms all other methods or has statistical power that is very close to the most powerful method with which it has been compared.
| S-EPMC7480017 | biostudies-literature
Project description:Haplotypes of closely linked single-nucleotide polymorphisms (SNPs) potentially offer greater power than individual SNPs to detect association between genetic variants and disease. We present a novel approach for association mapping in which density-based clustering of haplotypes reduces the dimensionality of the general linear model (GLM)-based score test of association implemented in the HaploStats software (Schaid et al. [2002] Am. J. Hum. Genet. 70:425-434). A flexible haplotype similarity score, a generalization of previously used measures, forms the basis, for grouping haplotypes of probable recent common ancestry. All haplotypes within a cluster are assigned the same regression coefficient within the GLM, and evidence for association is assessed with a score statistic. The approach is applicable to both binary and continuous trait data, and does not require prior phase information. Results of simulation studies demonstrated that clustering enhanced the power of the score test to detect association, under a variety of conditions, while preserving valid Type-I error. Improvement in performance was most dramatic in the presence of extreme haplotype diversity, while a slight improvement was observed even at low diversity. Our method also offers, for binary traits, a slight advantage in power over a similar approach based on an evolutionary model (Tzeng et al. [2006] Am. J. Hum. Genet. 78:231-242).
| S-EPMC2952426 | biostudies-literature
Project description:BackgroundClustering of observations is a common phenomenon in epidemiological and clinical research. Previous studies have highlighted the importance of using multilevel analysis to account for such clustering, but in practice, methods ignoring clustering are often employed. We used simulated data to explore the circumstances in which failure to account for clustering in linear regression could lead to importantly erroneous conclusions.MethodsWe simulated data following the random-intercept model specification under different scenarios of clustering of a continuous outcome and a single continuous or binary explanatory variable. We fitted random-intercept (RI) and ordinary least squares (OLS) models and compared effect estimates with the "true" value that had been used in simulation. We also assessed the relative precision of effect estimates, and explored the extent to which coverage by 95% confidence intervals and Type I error rates were appropriate.ResultsWe found that effect estimates from both types of regression model were on average unbiased. However, deviations from the "true" value were greater when the outcome variable was more clustered. For a continuous explanatory variable, they tended also to be greater for the OLS than the RI model, and when the explanatory variable was less clustered. The precision of effect estimates from the OLS model was overestimated when the explanatory variable varied more between than within clusters, and was somewhat underestimated when the explanatory variable was less clustered. The cluster-unadjusted model gave poor coverage rates by 95% confidence intervals and high Type I error rates when the explanatory variable was continuous. With a binary explanatory variable, coverage rates by 95% confidence intervals and Type I error rates deviated from nominal values when the outcome variable was more clustered, but the direction of the deviation varied according to the overall prevalence of the explanatory variable, and the extent to which it was clustered.ConclusionsIn this study we identified circumstances in which application of an OLS regression model to clustered data is more likely to mislead statistical inference. The potential for error is greatest when the explanatory variable is continuous, and the outcome variable more clustered (intraclass correlation coefficient is ≥ 0.01).
| S-EPMC8265092 | biostudies-literature
Project description:Subspecies nomenclature systems of pathogens are increasingly based on sequence data. The use of phylogenetics to identify and differentiate between clusters of genetically similar pathogens is particularly prevalent in virology from the nomenclature of human papillomaviruses to highly pathogenic avian influenza (HPAI) H5Nx viruses. These nomenclature systems rely on absolute genetic distance thresholds to define the maximum genetic divergence tolerated between viruses designated as closely related. However, the phylogenetic clustering methods used in these nomenclature systems are limited by the arbitrariness of setting intra and intercluster diversity thresholds. The lack of a consensus ground truth to define well-delineated, meaningful phylogenetic subpopulations amplifies the difficulties in identifying an informative distance threshold. Consequently, phylogenetic clustering often becomes an exploratory, ad hoc exercise. Phylogenetic Clustering by Linear Integer Programming (PhyCLIP) was developed to provide a statistically principled phylogenetic clustering framework that negates the need for an arbitrarily defined distance threshold. Using the pairwise patristic distance distributions of an input phylogeny, PhyCLIP parameterizes the intra and intercluster divergence limits as statistical bounds in an integer linear programming model which is subsequently optimized to cluster as many sequences as possible. When applied to the hemagglutinin phylogeny of HPAI H5Nx viruses, PhyCLIP was not only able to recapitulate the current WHO/OIE/FAO H5 nomenclature system but also further delineated informative higher resolution clusters that capture geographically distinct subpopulations of viruses. PhyCLIP is pathogen-agnostic and can be generalized to a wide variety of research questions concerning the identification of biologically informative clusters in pathogen phylogenies. PhyCLIP is freely available at http://github.com/alvinxhan/PhyCLIP, last accessed March 15, 2019.
| S-EPMC6573476 | biostudies-literature
Project description:True linear relationships are rare in clinical data. Despite this, linearity is often assumed during analyses, leading to potentially biased estimates and inaccurate conclusions. In this introductory paper, we aim to first describe-in a non-mathematical manner-how to identify non-linear relationships. Various methods are then discussed that can be applied to deal with non-linearity, including transformations, polynomials, splines and generalized additive models, along with their strengths and weaknesses. Finally, we illustrate the use of these methods with a practical example from nephrology, providing guidance on how to report the results from non-linear relationships.
| S-EPMC11852331 | biostudies-literature
Project description:Metagenomic datasets contain billions of protein sequences that could greatly enhance large-scale functional annotation and structure prediction. Utilizing this enormous resource would require reducing its redundancy by similarity clustering. However, clustering hundreds of millions of sequences is impractical using current algorithms because their runtimes scale as the input set size N times the number of clusters K, which is typically of similar order as N, resulting in runtimes that increase almost quadratically with N. We developed Linclust, the first clustering algorithm whose runtime scales as N, independent of K. It can also cluster datasets several times larger than the available main memory. We cluster 1.6 billion metagenomic sequence fragments in 10 h on a single server to 50% sequence identity, >1000 times faster than has been possible before. Linclust will help to unlock the great wealth contained in metagenomic and genomic sequence databases.
| S-EPMC6026198 | biostudies-literature
Project description:Aryl CH hydrogen bonds (HBs) are now commonly recognized as important factors in a number of fields, including molecular biology, stereoselective catalysis, and anion supramolecular chemistry. As the utility of CH HBs has grown, so to has the need to understand the structure-activity relationship for tuning both their strength and selectivity. Although there has been significant computational effort in this area, an experimental study of the substituent effects on CH HBs has not been previously undertaken. Herein we disclose a systematic study of a single CH HB by using traditional urea donors as directing groups in a supramolecular binding cavity. Experimentally determined association constants are examined by a combination of computational (electrostatic potential) and empirical (σm and σp) values for substituent effects. The dominance of electrostatic parameters, as observed in a computational DFT study, is consistent with current CH HB theory; however, a novel anion dependence of the substituent effects is revealed in solution.
| S-EPMC4977030 | biostudies-literature
Project description:There is strong evidence showing that joint analysis of multiple phenotypes in genome-wide association studies (GWAS) can increase statistical power when detecting the association between genetic variants and human complex diseases. We previously developed the Clustering Linear Combination (CLC) method and a computationally efficient CLC (ceCLC) method to test the association between multiple phenotypes and a genetic variant, which perform very well. However, both of these methods require individual-level genotypes and phenotypes that are often not easily accessible. In this research, we develop a novel method called sCLC for association studies of multiple phenotypes and a genetic variant based on GWAS summary statistics. We use the LD score regression to estimate the correlation matrix among phenotypes. The test statistic of sCLC is constructed by GWAS summary statistics and has an approximate Cauchy distribution. We perform a variety of simulation studies and compare sCLC with other commonly used methods for multiple phenotype association studies using GWAS summary statistics. Simulation results show that sCLC can control Type I error rates well and has the highest power in most scenarios. Moreover, we apply the newly developed method to the UK Biobank GWAS summary statistics from the XIII category with 70 related musculoskeletal system and connective tissue phenotypes. The results demonstrate that sCLC detects the most number of significant SNPs, and most of these identified SNPs can be matched to genes that have been reported in the GWAS catalog to be associated with those phenotypes. Furthermore, sCLC also identifies some novel signals that were missed by standard GWAS, which provide new insight into the potential genetic factors of the musculoskeletal system and connective tissue phenotypes.
| S-EPMC9975197 | biostudies-literature
Project description:Through an understanding of the conformational preferences of the polyketide natural product (-)-zampanolide, and the structural motifs that control these preferences, we developed a linear zampanolide analogue that exhibits potent cytotoxicity against cancer cell lines. This discovery provides a set of three structural handles for further structure-activity relationship (SAR) studies of this potent microtubule-stabilizing agent. Moreover, it provides additional evidence of the complex relationship between ligand preorganization, conformational flexibility, and biological potency. In contrast to medicinal chemistry dogma, these results demonstrate that increased overall conformational flexibility is not necessarily detrimental to protein binding affinity and biological activity.
| S-EPMC10615712 | biostudies-literature
Project description:Sorption isotherm data were determined for a set of 8 aromatic organic compounds with varying physical chemical properties in three soils with organic matter of differing quantity and composition. The primary goals of this study were to test single and multiparameter linear free energy relationships on their ability to predict the observed sorption behavior on different types of natural sorbents across a range of solutes and concentrations and to relate the accuracy of the predictions to sorbent and solute structural descriptors. Organic carbon normalized sorption coefficients (K(OC)) predicted using both single and multiparameter LFERs were in good agreement with experimental data obtained at the highest tested aqueous concentrations (average deviation less than 0.1 log units over all solutes and sorbents), but deviations were more substantial (0.59-0.65 log units) at the lowest tested concentrations. For chlorinated benzenes there was a significant correlation between experiment-prediction discrepancies and the aromatic content of the soil organic matter measured by (13)C NMR, and the magnitude of the effect was similar to that observed previously for dissolved organic matter.
| S-EPMC2941804 | biostudies-literature