This github project contains a generic snakemake pipeline for PopIns2. It is supposed to facilitate your PopIns2 run and provides examples for a streamlined workflow. Users are encouraged to freely use and modify the code.

Download this PopIns2 snakeproject via

git clone https://github.com/Krannich479/PopIns2_snakeproject.git

The top folder PopIns2_snakeproject will be referred as WORK_DIR. This guide assumes that every instruction will be executed from within the WORK_DIR unless stated otherwise. Adjust the variables in scripts/snake_config.yaml (see bottom of this section for more details) to your system environment and create some, for this particular snakemake pipeline, required subfolders via

mkdir unmapped
mkdir results

Create a subfolder per sample within the unmapped folder and copy (or rather link!) the mapped reads and their index into the respective sample folder. Your initial folder structure should then look like

$ tree .
.
├── README.md
├── results
├── scripts
│   ├── snake_config.yaml
│   └── Snakefile
└── unmapped
    ├── sample_01
    │   ├── sample_01.bam
    │   └── sample_01.bam.bai
    ├── sample_02
    │   ├── sample_02.bam
    │   └── sample_02.bam.bai
    └── sample_03
        ├── sample_03.bam
        └── sample_03.bam.bai

Note that in the default setup of this pipeline the sample subfolders unmapped/<sample_##> have the same name as the mapping files unmapped/<sample_##>/<sample_##>.bam.

The variables in the snake_config.yaml have to adjusted to your system/environment, i.e. for the dafault setup of this snakemake project the

• POPINS2_BIN variable must point to the popins2 binary
• REFERENCE variable must point to the reference file the samples got mapped to
• WORK_DIR variable must point to the PopIns2_snakeproject folder
• SAMPLES variable must contain a list of sample names, here, the names of the subfolders within {WORK_DIR}/unmapped. If your BAM file names differ from the subfolder names you have to adjust the BAM_FILE positional argument of the popins2 assemble call

The DAG above displays the workflow using three samples (HG00448, HG00449, HG00450). Each module, except for the genotype module, directly reports its output to the all rule and provides the input for the next successive module of PopIns2.