Dockerized mvMAPIT

In order to enable users to work with mvMAPIT without having to install all dependencies and libraries on their local, we provide a docker build with all dependencies and an installed R package mvMAPIT. Learn how to build your own version of the docker image.

Docker Setup

Follow the official guide to learn how to Get Docker. This is required for being able to follow this tutorial.

Build an Image with mvMAPIT

The github repository of mvMAPIT already comes with a Dockerfile. To build the image, clone the github repository and run the following commands.

cd mvMAPIT
docker build -t mvmapit .

This will produce an image named mvmapit that contains Rstudio, mvMAPIT, and all dependencies.

Run the mvMAPIT Image

With a local copy of the docker image mvmapit available, run the following code.

docker run --rm -ti \
    -e DISABLE_AUTH=true \
    -p 8787:8787 \
    --name my_container \
    mvmapit

This will start the docker container that serves an RStudio application at localhost:8787. In this container, mvMAPIT is already installed and can be imported and run in the R console via the following code.

library(mvMAPIT)
mvmapit(t(simulated_data$genotype[1:100,1:10]),
        t(simulated_data$trait[1:100,]),
        cores = 2, logLevel = "DEBUG")

## 2026-05-23 08:21:46.775076 DEBUG:mvmapit:Running in normal test mode.
## 2026-05-23 08:21:46.781651 DEBUG:mvmapit:Genotype matrix: 10 x 100
## 2026-05-23 08:21:46.78755 DEBUG:mvmapit:Phenotype matrix: 2 x 100
## 2026-05-23 08:21:46.788206 DEBUG:mvmapit:Number of zero variance variants: 0
## 2026-05-23 08:21:46.788901 DEBUG:mvmapit:Genotype matrix after removing zero variance variants: 10 x 100
## 2026-05-23 08:21:46.789356 DEBUG:mvmapit:Scale X matrix appropriately.
## 2026-05-23 08:21:46.790221 INFO:mvmapit:Running normal C++ routine.
## 2026-05-23 08:21:46.794405 DEBUG:mvmapit:Calculated mean time of execution. Return list.

## $pvalues
## # A tibble: 30 × 3
##    id        trait         p
##    <chr>     <chr>     <dbl>
##  1 snp_00001 p_01*p_01 0.435
##  2 snp_00001 p_02*p_01 0.476
##  3 snp_00001 p_02*p_02 0.909
##  4 snp_00002 p_01*p_01 0.510
##  5 snp_00002 p_02*p_01 0.606
##  6 snp_00002 p_02*p_02 0.326
##  7 snp_00003 p_01*p_01 0.452
##  8 snp_00003 p_02*p_01 0.536
##  9 snp_00003 p_02*p_02 0.140
## 10 snp_00004 p_01*p_01 0.688
## # ℹ 20 more rows
## 
## $pves
## # A tibble: 30 × 3
##    id        trait          PVE
##    <chr>     <chr>        <dbl>
##  1 snp_00001 p_01*p_01  0.0615 
##  2 snp_00001 p_02*p_01  0.0663 
##  3 snp_00001 p_02*p_02 -0.00451
##  4 snp_00002 p_01*p_01 -0.0215 
##  5 snp_00002 p_02*p_01 -0.0239 
##  6 snp_00002 p_02*p_02 -0.0287 
##  7 snp_00003 p_01*p_01  0.0579 
##  8 snp_00003 p_02*p_01  0.0493 
##  9 snp_00003 p_02*p_02 -0.0400 
## 10 snp_00004 p_01*p_01 -0.0146 
## # ℹ 20 more rows
## 
## $duration
##       process duration_ms
## 1         cov         NaN
## 2 projections         NaN
## 3   vectorize         NaN
## 4           q         NaN
## 5           S         NaN
## 6          vc         NaN
## 
## $vc
## # A tibble: 30 × 4
##    id        trait           VC     SE
##    <chr>     <chr>        <dbl>  <dbl>
##  1 snp_00001 p_01*p_01  0.0724  0.0928
##  2 snp_00001 p_02*p_01  0.0393  0.0551
##  3 snp_00001 p_02*p_02 -0.00490 0.0430
##  4 snp_00002 p_01*p_01 -0.0249  0.0378
##  5 snp_00002 p_02*p_01 -0.0145  0.0282
##  6 snp_00002 p_02*p_02 -0.0318  0.0323
##  7 snp_00003 p_01*p_01  0.0693  0.0921
##  8 snp_00003 p_02*p_01  0.0304  0.0491
##  9 snp_00003 p_02*p_02 -0.0442  0.0300
## 10 snp_00004 p_01*p_01 -0.0177  0.0440
## # ℹ 20 more rows