Getting started¶
The configuration file¶
The workflow comes with a configuration file config/config.yaml. You can
either modify this file directly or create a new config file and point snakemake
to it with snakemake --configfile <your-config-file>.yaml. The configuration
is validated
with a schema found under workflow/schemas/config.schema.yaml which is also
used to set default values. Specifying a configfile with --configfile extends
and overwrites the default settings so that any configfile you specify only
needs to contain the parameters you want to change.
For example, the default config file begins with:
sample_list: config/samples.tsv
paths:
results: "results"
A new config file, named e.g. conf_test.yaml, and containing only the
lines:
paths:
results: "test"
will then use test/ as output directory for results when the workflow is run
as:
snakemake --use-conda --configfile conf_test.yaml -j 4
The sample list¶
The workflow requires you to supply a file with some information on your
samples. The path to this file is specified with the sample_list: parameter
in the configuration file.
As a minimum the file must contain the columns sample, unit and fq1.
Paired-end samples also require the fq2 column for the mate 2 read file.
A very basic sample file may look like this:
| sample | unit | fq1 | fq2 | |
|---|---|---|---|---|
| sample1 | 1 | examples/data/sample1_1_R1.fastq.gz | examples/data/sample1_1_R2.fastq.gz | |
| sample2 | 1 | examples/data/sample2_11_R1.fastq.gz | examples/data/sample2_11_R2.fastq.gz | |
| sample3 | 1 | examples/data/sample3_21_R1.fastq.gz | ||
The columns fq1 and fq2 (for paired-end data) specify the paths to fastq
files which will be used as input to the workflow.
Specifying assemblies¶
You may also include a column named assembly with names for assemblies to
create. The assembly field can be comma-separated entries specifying several
assemblies per sample/unit combination.
For instance, with a sample file like this:
| sample | unit | assembly | fq1 | fq2 |
|---|---|---|---|---|
| sample1 | 1 | sample1,all | examples/data/sample1_1_R1.fastq.gz | examples/data/sample1_1_R2.fastq.gz |
| sample2 | 1 | sample2,all | examples/data/sample2_11_R1.fastq.gz | examples/data/sample2_11_R2.fastq.gz |
| sample3 | 1 | sample3,all | examples/data/sample3_21_R1.fastq.gz |
a total of 4 assemblies will be generated:
- sample1, sample2 and sample3 with input only from each sample respectively, and
- all with input from all samples