Open AccessPhenoMan: phenotypic data exploration, selection, management and quality control for association studies of rare and common variants
Author(s) -
Biao Li,
Gao T. Wang,
Suzanne M. Leal
Publication year - 2013
Publication title -
bioinformatics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.599
H-Index - 390
eISSN - 1367-4811
pISSN - 1367-4803
DOI - 10.1093/bioinformatics/btt682
Subject(s) - covariate , selection (genetic algorithm) , missing data , computer science , consistency (knowledge bases) , data mining , genetic association , outlier , subtyping , trait , data quality , metric (unit) , computational biology , biology , single nucleotide polymorphism , machine learning , artificial intelligence , genetics , operations management , genotype , gene , programming language , economics
Next-generation sequencing and other high-throughput technology advances have promoted great interest in detecting associations between complex traits and genetic variants. Phenotype selection, quality control (QC) and control of confounders are crucial and can have a great impact on the ability to detect associations. Although there are programs to perform association analyses, e.g. PLINK and GenABEL, they cannot be used for comprehensive management and QC of phenotype data. To address this need PhenoMan was developed: to select individuals based on multiple phenotype criteria or population membership; control for missing covariate data; remove related individuals, duplicate samples and individuals with incorrect sex specification; recode primary traits and covariates; transform data; remove or winsorize outliers; select covariates for analysis; and create residuals. To ensure consistency and harmonization between analyses, a report is generated for every dataset. Summary statistics are also provided in graphical or text format. PhenoMan can be used for selection and manipulation of quantitative, disease and control data.