a snakemake based workflow for demultiplexing Illumina runs with bclconvert
automatically does our most-used application, RNASeq analysis with a lot of QC before and after mapping.
For best results, use Ensembl reference genome and annotation!
-
bclconvert/bcl2fastq2
-
interop_plots
-
checksum creation
-
fastqc, fastp
-
cutadapt
-
umi_tools
-
kraken2
-
blastn
-
qualimap
-
rseqc
-
samtools stats, coverage, plot
-
stringtie
-
prevent weird auto-reverse complement in read 3 if samplesheet created on the wrong Machine
-
STAR as main mapper
-
2 multiqc reports: simple + all analysis inclusive
-
auto-transfer of folder to other servers
-
gocryptfs
-
preseq
-
sortmerna
the workflow is actually 3 parts:
- demultiplexing, interop plot, checksums of untrimmed data
- fastqProcessing: all other parts, excluding
- RseQC: only active if rseqc_modules_active= True in config.yaml
git pull
edit config.yaml
bash runPipeline.sh