CRC atlas - velocyto nextflow pipeline

#TODO: documentation

1.1. Usage

Example for 10X data

$ nextflow run main.nf --input_csv <sampleSheet.csv> --gtf_file <ref_anno.gtf> --masked_repeats <rmsk.gtf> --out_dir <~/outdir/> --platform 10X -profile cluster,conda

Example for BD data

$ nextflow run main.nf --input_csv <sampleSheet.csv> --gtf_file <ref_anno.gtf> --masked_repeats <rmsk.gtf> --out_dir <~/outdir/> --platform BD -profile cluster,conda

<Arguments>

--input_csv = path to the input CSV file (see below for details)

--gtf_file = path to GTF file (used for the creation of the BAM files)

--masked_repeats = path to the masked repeats file

--out_dir = path to the output directory

--publish_dir_mode = mode of publishDir directive (default: "copy")

--platform = [str: 10X/BD] choose the platform the samples were sequenced on (default: "")

--convert_loom = [str: true/false] renames the barcodes in the loom file according to the <cohort-sampleID-barcode> convention - this module is tested only on ICBI's CRC atlas datasets (default: "false")

--custom_loom_dir = if --confert_loom true specify the ouput dir for the converted loom files

Notes

**Outpur directory

The pipeline will output the results in the /path/to/Sample.bam directory by default. You need to specify a directory through --out_dir otherwise.

CSV file (default settings)

study,sample_id,bam,bcl
cohort_1,sample_1,/path/to/Sample1_possorted.bam,/path/to/Sample1_barcodes.tsv
cohort_2,sample_2,/path/to/Sample2_possorted.bam,/path/to/Sample2_barcodes.tsv

• If you choose the --platform 10X you need to change the input CSV file to specify the cellranger directories instead of specific bam files (not tested yet)

CSV file (run_10x flag)

study,sample_id,bam
cohort_1,sample_1,/path/to/Sample1/
cohort_2,sample_2,/path/to/Sample2/