README.md

In aballou16/geneticriskR: Standard Analyses for Polygenic Risk Score Data for PRSice-2 output

geneticriskR

Package Aims

The goal of geneticriskR is to provide standard formatting, statistical analyses and plotting for polygenic risk score data generated using PRSice-2 software. Because recent genetic risk studies have shown a great deal of inconsistency, this package will aid the reproducibility of future studies.

Installation

You can install the development version of geneticriskR from Github with:

# install.packages("devtools")
devtools::install_github("aballou16/geneticriskR")

Inspiration

PRS research has seen significant backlash over the past decade. Much of this controversy is the result of inconsistent, unreproducible results. While sharing genetic data is difficult (due to privacy and patient protection laws), using geneticriskR will allow for other researchers to understand and replicate the analyses that formed the study.

Example(s)

This is a basic example which demonstrates how to run a simple logistic regression model using an example output file from PRSice-2 called “test_data.”

library(geneticriskR)
simple_mod <- simple_logistic_reg(test_data, "hypertension")
simple_mod
#> 
#> Call:  glm(formula = full_table[[phenotype]] ~ prs, family = "binomial", 
#>     data = full_table)
#> 
#> Coefficients:
#> (Intercept)          prs  
#>    -1.14120      0.05571  
#> 
#> Degrees of Freedom: 3299 Total (i.e. Null);  3298 Residual
#> Null Deviance:       3655 
#> Residual Deviance: 3654  AIC: 3658

This is another example which demonstrates how to run roc/auc analysis on the model generated above.

proc_analysis(test_data, "hypertension", simple_mod)
#> Setting levels: control = 0, case = 1
#> Setting direction: controls < cases

#> 
#> Call:
#> roc.default(response = full_table[[phenotype]], predictor = my_model$fitted.values,     plot = TRUE)
#> 
#> Data: my_model$fitted.values in 2500 controls (full_table[[phenotype]] 0) < 800 cases (full_table[[phenotype]] 1).
#> Area under the curve: 0.5138

This is an example of one of the standard plots included in most PRS analyses:a side by side box plot comparing scores for cases and conrols.

compare_boxplots(test_data, "hypertension", "prs")
#> cases = 1, controls = 0

aballou16/geneticriskR documentation built on Dec. 27, 2019, 3:39 p.m.