You will need a kallisto index file, an annotations.gtf file and the
input fastq files. The output is a set of QC reports and a count
table.
The internal scripts expect the following folder structure
/index/index.kidx: the kallisto index file./index/annotations.gtf: a annotations file used to convert transcript counts to gene counts./input/XXXXX,/input/YYYYY, etc. the "runs" to be used in the alignment. In practice these are the folders containingfastq.gzfiles that are interpreted as samples.
You can specify a config file for the pipeline by mounting a
/config.yaml file (via e.g. - my_config.yaml:/config.yaml:ro).
The config can be used to select the samples to process and change the
kallisto parameters. For example, to enable pseudobam generation, run
kallisto in single-end mode and process only samples [4141,4136] you
can use the following config file
kallisto:
mode: "single"
pseudobam: true
threads: 10
readlenght: 75
readstd: 1
samples: [4141,4136]
The allowed values for the mode parameter are
mode=["single","paired","auto"]. To run the alignment for all
samples use samples=[].