BulkLMMSysimage includes instructions for generating the BulkLMM.jl sysimage along with the code for the command-line functions.

Julia employs a just-in-time (JIT) compiler, which means that computer code is compiled during program execution, or at runtime, rather than before. When a function is called for the first time, Julia compiles it, which can be a time-consuming process. However, subsequent calls to that function within the same session will use the pre-compiled version for faster execution. If Julia is restarted, the compiled work is lost. To minimize this delay, one option is to use PackageCompiler.jl, which provides three different workflows. We decided to follow the sysimage workflow.

A sysimage is a file where we can save our loaded packages and compiled functions that we pass to Julia at the startup. In general, we cannot relocate a sysimage to another machine. The sysimage will only work on the machine where we created it.
We can consider a sysimage file as a Julia session serialized to a file. When we start Julia with a sysimage, the stored Julia session is deserialized and loaded. By using the sysimage file, we expect that this "deserialization" will be faster than having to reload packages and recompile code from scratch.

Reference: PackageCompiler.jl documentation

To generate the sysimage start Julia in this directory.

BulkLMMSysimage$ julia

To make the number of threads available to Julia you can start julia specifying the number of threads such as:

BulkLMMSysimage$ julia --threads 4

Then, run the following command in Julia to build the BulkLMM_API.so sysimage:

julia> include("run_compilation.jl")

Once the sysimage is created, we can use our pre-compiled Julia version by using simply the following command line, assuming that we are in the BulkLMM_API directory:

julia --project --sysimage=<sysimage_path> <scan_function_path > <phenotype_file_path> <genotype_file_path> <kinship_file_path> <output_lod_file_path> <number_of_permutation>

• Without permutations:

julia --project --sysimage=bin/BulkLMM_API.so bin/single_trait_scan.jl data/preprocessed/single_trait.csv data/preprocessed/geno.csv data/preprocessed/kinship.csv data/output/lod_1.csv

• With permutations:

julia --project --sysimage=bin/BulkLMM_API.so bin/single_trait_scan.jl data/preprocessed/single_trait.csv data/preprocessed/geno.csv data/preprocessed/kinship.csv data/output/lod_1.csv 1000 

NOTES: In the case we use a package that belongs to the registry or is included in the Project.toml, we just have to specify the path of the sysimage and/or the script; the --project flag should be removed.