sRNAmake

Table of contents

1. Introduction
2. Prepare inputs
3. Launch the pipeline
4. The outputs
5. citation
6. licence

Introduction :

sRNAmake is a pipeline of analyse of smallRNA using ShortStack and Deseq2.

Prepare config.yaml :

• inputs :

    'directories':
        'fastq_dir': "/path/to/directory/fastq"       # samples dir with fastq
        'out_dir': "/path/to/directory/output"       # Out director
    'files':
        'reference': "/path/to/file/ref.fasta"
        'annotation': "/path/to/file/ref.gtf"
        'miRNA_gff' : '/path/to/file/miRBase.gff3'
        'sample_info': "/path/to/file/sample_info.txt"
        'de_comparisons_file': "/path/to/file/treatmentsComparisons.csv"
        'filter_gff': "/path/to/file/filterGFF.gff3"

• Paste the directory of your fastq to analyse in fastq_dir
• Paste the path of the output directory in out_dir. If the directory does not exists it will be created.
• the file reference is the reference genome file in fasta format.
• the file annotation is the gff3 or gtf for the reference genome file.
• miRNA_gff is the annotation for the miRNA from mirbase for out organism.
• sample info is a descripting file of the fastq directory. It should be a tabulated file that have exactly the same header as indicated on the example below. It contains, the path of each file (Filename), the sample name (SampleName) witch will be reused in the rest of the pipeline (should be unique foreach sample), the Treatment and the experiment.

    Filename	SampleName	Treatment	Experiment
/path/to/file/fastq/1_cutadapt.fastq.gz	1_cutadapt	RYMV	R1
/path/to/file/fastq/2_cutadapt.fastq.gz	2_cutadapt	MOCK	R1
/path/to/file/fastq/3_cutadapt.fastq.gz	3_cutadapt	RYMV	R2
/path/to/file/fastq/4_cutadapt.fastq.gz	4_cutadapt	MOCK	R2
/path/to/file/fastq/5_cutadapt.fastq.gz	5_cutadapt	RYMV	R3
/path/to/file/fastq/5_cutadapt.fastq.gz	6_cutadapt	MOCK	R3

• The de_comarisons_file is a semicolon separated csv file which indicate every comparison to do durin the differential expression step of the pipeline.

condA;condB
RYMV;MOCK

• The filter_gff file is a possibility to use a different gff for the DE analysis step.

• parameters :

Tools parameters can be changed in the config.yaml file.

PARAMS:
    FASTP:
        options: "--overrepresentation_analysis --length_required 18 --length_limit 30"
        #options: "-A -Q -L -G --overrepresentation_analysis"
    BWA_ALN:
        options: ""
    SHORTSTACK:
        dicermin: 20
        dicermax: 24
        mincov: 1rpm
        pad: 75
        foldsize: 300
        strand_cutoff: 0.8
    DE_ANALYSIS:
        # feature filtering based on total read counts
        minRowSumTreshold: 30
        # DESeq2 modeling with GLM
        variableOfInterest: "Treatment"
        batch: "Experiment"
        locfunc: "shorth" # "median"
        fitType: "parametric"
        sizeFactEstimMethod: "iterate" # "ratio" # "poscounts"
        pAdjustMethod: "BH"
        cooksCutoff: "TRUE"
        independentFiltering: "FALSE"
        lfcThreshold: 1
        altHypothesis: "greaterAbs"
        alpha: 0.05
        # DESeq2 VSR transformation method
        typeTrans: "rlog"

tips:

• To make fewer and bigger clusters, use a mincov bigger for ShortStack parameters

• To be less stringent for the DE analysis, decrease lfcThreshold value and increase alpha value

• ###The containers:

SINGULARITY:
    "MAIN": "Containers/Singularity.R.simg"

This pipeline works with a system of containers. It uses a main singularity container (given in the directory Container with the pipeline) and several conda containers (in the directory envs/ of the pipeline).

Do not move the conda containers from the envs directory.

However, you should build the Singularity container in sudo using the command below and indicate the simg path in the config.yaml.

singularity build /path/to/simg/out.simg /path/to/def/in.def

How to lauch the pipeline ?

LOCAL

snakemake --use-singularity --use-conda --nolock --cores -p --verbose -s sRNA_pipeline.smk --latency-wait 500 --keep-going --restart-times 1 --rerun-incomplete  --configfile config/config.yaml

LOCAL on a node of itrop cluster

module load system/singularity/3.3.0
module load system/python/3.7.2
snakemake --use-singularity --use-conda --nolock --cores -p --verbose -s sRNA_pipeline.smk --latency-wait 500 --keep-going --restart-times 1 --rerun-incomplete  --configfile config/config.yaml

CLUSTER (With Slurm)

Sbatch Laucher_slurm.sh

• the file config/cluster_config_slurm.yaml:

You can change the way to lauch each rule. If nothing is indicated for a specific rule, it will launch itself with the default.

__default__:
    cpus-per-task : 2
    ntasks : 1
    mem-per-cpu : '4'
    partition : "normal"
    output : 'logs/stdout/{rule}/{wildcards}'
    error : 'logs/error/{rule}/{wildcards}'

bwa_index:
    ntasks : 1
    cpus-per-task : 4
    mem-per-cpu : '4'
    partition : "normal"
    output : 'logs/stdout/{rule}/{wildcards}'
    error : 'logs/error/{rule}/{wildcards}'

The outputs

1_QC
2_mapping_sRNA
3_merge_bam_sRNA
4_ShortStack
5_sRNA_loci_DE_analysis
6_MULTIQC
LOGS
report
slurm_log

• 1_QC: contains quality checking of fastq before and after filtering with fastp.
• 2_mapping_sRNA: raw bam of the alignments of each fastq on the reference
• 3_merge_bam_sRNA : merging of all the bam in one, and bam merging by condition
• 4_ShortStack: outputs of the Shortstack tool
• 5_sRNA_loci_DE_analysis: results of the differential analysis
• 6_MULTIQC : merging all the quality files, QC of fastq and QC of mapping.
• LOGS: logs for all the tools used.

Citation

@Authors:

Sebastien Cunnac (IRD) and Aurore Comte (IRD)

Thanks

The authors acknowledge the IRD i-Trop HPC (South Green Platform) at IRD Montpellier for providing HPC resources that have contributed to this work. https://bioinfo.ird.fr/ - http://www.southgreen.fr

License

Licencied under CeCill-C (http://www.cecill.info/licences/Licence_CeCILL-C_V1-en.html) and GPLv3 Intellectual property belongs to IRD and authors.