| N3finemapping {susieR} | R Documentation |
Simulated Fine-mapping Data with Three Effect Variables.
Description
The data-set contains a matrix of 574 individuals and 1,001 variables. These variables are real-world genotypes centered and scaled, and therefore retains the correlation structure of variables in the original genotype data. 3 out of the variables have non-zero effects. The response data is generated under a multivariate linear regression model. See Wang et al (2020) for more details.
Format
N3finemapping is a list with the following elements:
- X
N by P variable matrix of centered and scaled genotype data.
- chrom
Chromomsome of the original data, in hg38 coordinate.
- pos
Chromomosomal positoin of the original data, in hg38 coordinate. The information can be used to compare impact of using other genotype references of the same variables in susie_rss application.
- true_coef
The simulated effect sizes.
- residual_variance
The simulated residual covariance matrix.
- Y
The simulated response variables.
- allele_freq
Allele frequency of the original genotype data.
- V
Prior covariance matrix for effect size of the three non-zero effect variables.
References
G. Wang, A. Sarkar, P. Carbonetto and M. Stephens (2020). A simple new approach to variable selection in regression, with application to genetic fine-mapping. Journal of the Royal Statistical Society, Series B doi: 10.1101/501114.
Examples
data(N3finemapping)