Reproducing our analysis is computationally intensive, therefore we recommend to run the pipeline on a HPC.
The Nextflow alignment pipelines generate numerous temporary files stored within work directories. You can clear these after each alignment without affecting the outputs required for downstream analysis. Without clearing the temporary files, we recommend that you have at minimum ~4TB available storage, 8CPUs and 64GB RAM.
In order to reproduce our analysis, you will need to:
Important! All shell scripts are to be executed from the base directory of the project!
In order to reproduce our analysis, you will first need to download our otic-repregramming GitHub repository. To do this run:
git clone https://github.com/alexthiery/otic-reprogramming
NGS files (fastqs) are available on GEO under the accession GSE168089
To download the Galgal6 genome from Ensembl, run:
rsync -av rsync://ftp.ensembl.org/ensembl/pub/release-97/fasta/gallus_gallus/dna/Gallus_gallus.GRCg6a.dna.toplevel.fa.gz <LOCAL_PATH>
To download the Galgal6 gtf from Ensembl, run:
rsync -av rsync://ftp.ensembl.org/ensembl/pub/release-97/gtf/gallus_gallus/Gallus_gallus.GRCg6a.97.gtf.gz <LOCAL_PATH>
Once the genome files have been downloaded, they need to be unzipped before running the alignment.
Configuration properties and file paths are passed to Nextflow via configuration files.
In order to re-run the alignments and downstream analysis for this project, you will need to create a custom configuration file which contains:
Below is a copy of the config file used to run the pipeline at The Francis Crick Institute HPC. Given that this alignment was run on a HPC, we configure Nextflow to run with Singularity instead of Docker.
#!/usr/bin/env nextflow
singularity {
enabled = true
autoMounts = true
docker.enabled = false
}
process {
executor = 'slurm'
}
params {
max_memory = 224.GB
max_cpus = 32
max_time = 72.h
fasta = "/camp/home/thierya/working/genomes/galgal6/Gallus_gallus.GRCg6a.dna.toplevel.fa"
gtf = "/camp/home/thierya/working/genomes/galgal6/Gallus_gallus.GRCg6a.97.gtf"
}
Further information on Nextflow configuration can be found here.
After you have saved your custom configuration file, you are ready to align the data!
When running the NF-core alignments, the pipeline may be interupted as the relevant docker images are downloaded. If this happens, simply re-run the shell script and the pipeline will continue from where it ended.
Our ChIPseq data was aligned using the NF-core ChIPseq v1.2.0 pipeline.
Sample FASTQ paths are provided via a samplesheet csv. The template csv used to run this analysis can be found here.
After saving the required files, you can run the alignment with the following shell script.
export NXF_VER=20.07.1
nextflow pull nf-core/chipseq
nextflow run nf-core/chipseq \
-r 1.2.0 \
-c <PATH_TO_CONFIG> \
--input <PATH_TO_SAMPLESHEET.CSV> \
--macs_gsize 1.05e9 \
--skip_diff_analysis \
--outdir output/NF-ChIP_alignment \
--email <INSERT_EMAIL_ADDRESS> \
-resume
Our ATACseq data was aligned using the NF-core ATACseq v1.2.0 pipeline.
Sample FASTQ paths are provided via a samplesheet csv. The template csv used to run this analysis can be found here.
After saving the required files, you can run the alignment with the following shell script.
export NXF_VER=20.07.1
nextflow pull nf-core/atacseq
nextflow run nf-core/atacseq \
-r 1.2.0 \
-c <PATH_TO_CONFIG> \
--input <PATH_TO_SAMPLESHEET.CSV> \
--macs_gsize 1.05e9 \
--narrow_peak \
--skip_diff_analysis \
--outdir output/NF-ATAC_alignment \
--email <INSERT_EMAIL_ADDRESS> \
-resume
Our bulk RNAseq data were aligned using the NF-core RNAseq v2.0 pipeline.
We need to align two separate bulk RNAseq datasets.
Lmx1aE1 vs Sox3U3
First, we collected cells from putative otic and epibranchial fates using Lmx1aE1 and Sox3U3 enhancers respectively.
An example samplesheet csv can be found here.
Execute the following shell script to run the Lmx1a alignment.
export NXF_VER=20.07.1
nextflow pull nf-core/rnaseq
nextflow run nf-core/rnaseq \
-r 2.0 \
-c <PATH_TO_CONFIG> \
--aligner star \
--input <PATH_TO_LMX1A_SAMPLESHEET.CSV> \
--outdir output/NF-lmx1a_alignment \
--email <INSERT_EMAIL_ADDRESS> \
-resume
Sox8 overexpression
Next, we collected cells following Sox8 overexpression and compared relative to control cells.
Sox8 overexpression samples are double positive Sox8-mCherry/Lmx1a-EGFP, whilst control samples are labelled with constitutive mCherry/EGFP.
An example samplesheet csv can be found here
Execute the following shell script to run the Sox8 overexpression alignment.
export NXF_VER=20.07.1
nextflow pull nf-core/rnaseq
nextflow run nf-core/rnaseq \
-r 2.0 \
-c <PATH_TO_CONFIG> \
--aligner star \
--input <PATH_TO_SOX8_SAMPLESHEET.CSV> \
--outdir output/NF-sox8_alignment \
--email <INSERT_EMAIL_ADDRESS> \
-resume
Our Smartseq2 single cell RNAseq data was aligned using a custom DSL2 Nextflow pipeline. Details of the pipeline processes can be found here.
As with the NF-core pipelines above, our custom SmartSeq2 alignment pipeline passes the sample FASTQ files via a samplesheet csv. An example samplesheet can be found here.
As we have individual FASTQ files for each cell, it would be cumbersome to list each file individually. Instead we can pass an entire sample FASTQ directory which is then enumerated in Nextflow. For this to work, only the paths in column 2 ($data1) in the samplesheet csv should be edited.
Once you have saved your samplesheet, execute the following shell script to run the SmartSeq2 alignment.
export NXF_VER=20.07.1
nextflow run ./NF-smartseq2_alignment/main.nf \
-c <PATH_TO_CONFIG> \
--input <PATH_TO_SMARTSEQ2_SAMPLESHEET.CSV> \
--outdir output/NF-smartseq2_alignment \
-resume
We have built a custom Nextflow pipeline to integrate the outputs from the separate alignments.
After aligning the data, you will need to create a samplesheet csv containing the paths to all of the alignment output folders. You can find an example of this here.
To run the integrated downstream analysis execute the following shell script.
export NXF_VER=20.07.1
nextflow run ./NF-downstream_analysis/main.nf \
-c <PATH_TO_CONFIG> \
--input <PATH_TO_DOWNSTREAM_ANALYSIS_SAMPLESHEET.CSV> \
--outdir output/NF-downstream_analysis \
-resume
If you would like to interactively carry out downstream anlysis click here.