grg_pheno_sim

This is a code repository to simulates phenotypes on GRGs (genotype representation graphs). The simulator first simulates effect sizes based on the user's desired distribution model (a wide spectrum of options are provided, both for simulation of single and multiple causal mutations at a go), computes the genetic values by passing the effect sizes down the genotype representation graph, and then adds simulated environmental noise to obtain the final phenotypes for the individuals in the graph. Normalization of genetic values is provided as well, either prior to adding environmental noise or after noise is added, according to the user's desire. In addition, there is an option to use normalized genotypes. The simulator offers the simulation of binary phenotypes as well, in addition to simulation on multiple GRGs simultaneously. Finally, options to obtain standardized outputs for both effect sizes (.par files) and phenotypes (.phen files) are included as well.

The folder grg_pheno_sim contains all the primary source code for the simulator. The demos folder contains ipynb notebooks with sample uses and demomstrations of the different stages of the phenotype simulator. It also contains incremental verifications of outputs to ensure accurate simulation. The test_phenotype_sim folder contains a suite of test functions used in the demos.

Documentation can be found here.

Installation

Installing from pip

If you just want to use the tools offered by grg_pheno_sim then you can install via pip (from PyPi)

pip install grg_pheno_sim

Installing from source

1. Clone the repository
2. If you wish to install the package without any changes to source code, use pip install /path/to/grg_pheno_sim/ (this is for standard installation)
3. If you wish to install the package and modify the source code, use pip install -e /path/to/grg_pheno_sim/ (this is for development installation)

Usage

The demos folder contains a vast repository of use cases for the phenotype simulator, including sample outputs and standardized outputs commands (the output files themselves are excluded from the GitHub repo but can be easily obtained by running the appropriate notebook).