Pipeline for calling variants for Lutzoymia samples
The pipeline is organized into four stages:
- Alignment: trims reads, aligns reads to genome
- Haplotyping: Runs the GATK HaplotypeCaller on each BAM file
- Genotyping: Combines the resulting GVCFs and performs genotype calling
- Variant Filtering: Performs quality filtering, separates SNPs from indels
The pipeline expects to find a genome FASTA file and pairs of FASTQ files in the input directory. The name of the genome file can be set in the config.yaml file.
The other files are found by searching for pairs of *_1.fastq.gz. and *_2.fastq.gz in the input/samples directory.
Parameters for trimming and haplotyping can be set in the config.yaml file. The variant filtering parameters are currently hard-coded in variant_filtering.Snakefile, but they will be moved to config.yaml in the future. Parameters were taken primarily from the 1000 Anopheles genomes supplemental materials.
You can run the pipeline stages as follows:
$ snakemake --cores X --snakefile alignments.Snakefile run_pipeline
$ snakemake --cores X --snakefile haplotyping.Snakefile run_pipeline
$ snakemake --cores X --snakefile genotyping.Snakefile run_pipeline
$ snakemake --cores X --snakefile variant_filtering.Snakefile run_pipelinewhere X is the number of cores to use. The pipeline is set up to run on a single, high-memory machine.
Note that GATK version 4 will be needed. GATK 4 depends on JVM version 8, so that might also need to be installed.