Data management tasks in WISARD

In WISARD, mainly four kinds of data management tasks are possible.

1. Data retrieval - WISARD can utilize various kinds of data format that are widely accepted and used in several toolsets.
2. Data filtering - In WISARD, numerous filtering schemes can be applied to the dataset according to the subject of analyses.
3. Missing data treatment - WISARD supports several missing treatment policies including genotype, phenotype, covariate, and family structure.
4. Data reformatting - retrieved dataset can be split or reformatted into subsets or other formats, respectively, using WISARD.

Data types used in WISARD

WISARD is capable to handling various types of dataset, and those can be categorized like below figure.

Main dataset [top]

Main dataset is the core of analysis. Without main dataset, any analysis cannot be performed.

Genotype/dosage data

Pedigree structure data

Marker information data

Sample variable [top]

Samples in the dataset may have additional information other than the variables to be analyzed, and it may help to improve the result of analysis. Otherwise, since some file format cannot include an information other than genotype, interested trait(phenotype) or the other information must be supplied externally. WISARD calls this information as sample variables.

About the management of sample variable, please refer this section and this section.

Sample relatedness [top]

The genotypes of genetically closed samples, such as same population or relatives are highly related each other. In the analysis of dataset consists of such samples, their relationship need to be considered in the analysis. WISARD provides both of its computation and consideration in the analysis.

About the computation of sample relatedness and the way to retrieve it, please refer this section.