Genomics Platform

reNEW, CPR, CGEN, ICMM

Overview of pipelines offered by Genomics Platform

The following pipelines can be selected for the Genomics Platform to run:

in-house pipelines offered by Genomics Platform:

ATAC
CELL-RANGER
CHIP-CHOR-SCAR
CRISPRESSO2
CUTRUN-CUTTAG
FASTQ
MAGECK
MAGECK-BEAN
MAGECK-DRUGZ-BAGEL
RAW

nf-core pipelines offered by Genomics Platform

NF-CHIP (nf-core/chipseq)
NF-CUTANDRUN (nf-core/cutandrun)
NF-METHYL (nf-core/methylseq)
NF-RNA (nf-cpre/rnaseq)

ATAC

ATAC pipeline aligns reads in .fastq to a specified reference genome, removes chrM and blacklisted regions for mouse. Spike-in option is not built in by default, but alignment can be done on a spike-in genome separately.

Inputs: .fastq.gz files, a reference genome
Options: paired-end only
Outputs: indexed, sorted .bam files with chrM removed, .bw files
Software used: bowtie2, samtools, picard, multiQC, bedtools

CELL-RANGER

CELL-RANGER pipeline uses cellranger v.7.1.0 to demultiplex, count and aggregate the reads. Depending on library design (with or without cell hashing), further cellranger count + cellranger aggr or cellranger multi is used to create count matrixes for downstream analysis in Loupe browser, Seurat, etc. By default, the pipeline uses the pre-built reference transcriptome by 10X for mouse and human.

CHIP-CHOR-SCAR

CHORseq pipeline aligns reads in .fastq.gz to a specified reference genome, using dm6 as the spike-in genome.This pipeline is used to align reads for CHIP and CHOR assays. This pipeline can handle UMIs.

Inputs: .fastq files, a reference genome
Options: single-end, paired-end, with and without spike-in
Outputs: indexed, sorted .bam files
Software used: bowtie2, samtools, picard, multiQC, umi_tools, custom scripts

CHOR pipeline

CRISPRESSO2

CRISPRESSO2 pipeline is used for amplicon submissions and described here.

CUTRUN-CUTTAG

CUTRUN-CUTTAG pipeline is derived from the ATAC pipeline to handle reads produced by CUT&Run and Cut&Tag assays.

Inputs: .fastq.gz files, a reference genome
Options: paired-end only
Outputs: indexed, sorted .bam files, .bw files
Software used: bowtie2, samtools, picard, multiQC, bedtools, custom scripts

FASTQ

FASTQ pipeline demultiplexes raw images (.bcl) into .fastq.gz files according to samplesheet provided by the user. It can handle UMIs.

Please find the most updated samplesheets here

Inputs: raw .bcl files and a sample sheet
Outputs: .fastq files, multiQC report, md5sums for fastq files
Software used: bcl2fasta, multiQC

demux pipeline

MAGECK

This pipeline was developed with Emil Hertz (CPR) and tailored for specific CRISPR primers following Durocher’s lab protocol. If unaware, please ask us about it. The pipeline trims the reads (taking into account the staggers if present) and then run Mageck on the trimmed fastq with the library file indicated in the samplesheet.

MAGECK-BEAN

For CRISPR_reporter_library submissions, we typically run Mageck as well as BEAN. We run Mageck 3 times as follows:

guide on R1
rev-compl of guide on R2
insert on R1

MAGECK-DRUGZ-BAGEL

This pipeline starts with the Mageck pipeline explained above to compute the raw count matrix. After that, we run DrugZ on the raw counts. Also, we compute a normalised count matrix and run Bagel. For all these steps, we need you to email us the details of the experimental design amd which exact comparisons you need.

RNA (deprecated)

RNAseq pipeline aligns reads in .fastq to a specified reference genome using STAR. We currently offer running this pipeline for projects that need wrapping-up, but for the new projects we recommend using NF-RNA pipeline.

Inputs: .fastq.gz files
Outputs: indexed, sorted .bam files, read counts in .tsv format
Software used: STAR, multiQC

RNA pipeline

RAW

If you select RAW option, we will give you access to the run folder containing raw images (.bcl), so you can demultiplex the reads yourself.

NF-CORE PIPELINES USED AT GENOMICS PLATFORM

Apart from maintaining in-house pipelines, Genomics Platform can run some nf-core pipelines if the user requests it:

NF-CHIP (nf-core/chipseq)
NF-CUTANDRUN (nf-core/cutandrun)
NF-METHYL (nf-core/methylseq)
NF-RNA (nf-cpre/rnaseq)

We also highly recommend the users to try running these pipelines by themselves and are happy to provide support and training for running these pipeliens whenever a user requests help.

A profile for DANHEAD is available for nf-core pipelines nf-core/configs. Please see the pipeline use guidelines for nf-core pipelines here.

RUNNING OTHER COMMUNITY PIPELINES

You can always ask for help to run other community-maintained pipelines that are within nf-core and SnakePipes platforms. To run snakepipes on dangpu server run the following command and follow the instructions.

module load miniconda/latest snakePipes/2.5.4

The snakePipes workflows call SLURM jobs by default, so you do not need to run snakePipes workflows as slurm jobs. You can specify max CPUs with an option -j. Remember to not exceed the maximum allowed number of CPUs on DANHEAD during the work hours. You can read more on how to use DANHEAD HPC here.

AVAILABLE REFERENCE GENOMES

refgenie reference genome manager is available for accessing for pre-built reference genomes on dangpu server:

# load modules
module load dangpu_libs python/3.7.13 refgenie/0.12.1

# list genomes
refgenie list -g  GRCh38_ensembl

# find a path to the file of interest
# syntax: refgenie seek <genome>/<asset>:<tags>
refgenie seek  GRCh38_ensembl/bowtie2_index
refgenie seek  GRCh38_ensembl/blacklist:CUTANDRUN

refgenie documentation
more guidelines for dangpu server

Go back to the Genomics Platform home