🧬 Workflows to preprocess sequencing data

Warning To run this workflow it is necessary to have snakemake and Singularity installed on the computer.

Files needed to run the pipeline

• Raw fastq.gz files or SRA list

Map to reference

• Reference and gff file downloaded from ENSEMBL.

De novo assembly

• sample_file.txt and constrast_file.txt (drafts in the folder data) ➡️ it must be located on the input directory.

Running

First of all, set the file config.yaml with the name of your samples, directories and other settings.

Warning Do not forget of change the name of the samples and directories in the config.yaml file

Command to run:

snakemake -s <WORKFLOW_NAME> -j <N_OF_JOBS> --use-singularity --singularity-args "-B <DATA_DIRECTORY>"

It is also possible to create new workflows just combining rules by including in the main Snakefile.

Step-by-step

☑️ You have snakemake and singularity installed and running

☑️ You have the folder with the fastq files or the Sra Acc List

☑️ You have the Contrast and sample files (only if using de novo assembly)

☑️ Clone this repository on your computer

☑️ Edit config file with your pathways

☑️ Edit the config file path (1st line) on the file of the chosen pipeline

Download public data using SRA-tools

snakemake -s sra -j <N_OF_JOBS> --use-singularity --singularity-args "-B <DATA_DIRECTORY>"

These rules can be included in the workflow if you want to start it from the SRA list. It will download the files and perform the fastq-dump step to generate the fastq files.

• prefetch
• fastq-dump

Quality control

snakemake -s qC[SE/PE] -j <N_OF_JOBS> --use-singularity --singularity-args "-B <DATA_DIRECTORY>"

This workflow will create the quality control report with the raw data individually and also the merged report. It will trim adapters and bases with bad quality and then generates reports for the trimmed data.

• fastqc
• trimmomatic
• fastqc
• multiqc

RNAseq preprocessing

This workflow will perform the quality control steps and then, align the reads to a reference genome/transcriptome using STAR and creating the count matrix using RSEM.

snakemake -s preprocess[SE/PE] -j <N_OF_JOBS> --use-singularity --singularity-args "-B <DATA_DIRECTORY>"

• STAR
• RSEM

De novo transcriptome assembly

snakemake -s denovo[SE/PE] -j <N_OF_JOBS> --use-singularity --singularity-args "-B <DATA_DIRECTORY>"

Assembly

• trinity - assembly
• busco - checking quality
• cdhit - remove redundance

Annotation (Trinotate pipeline)

• TransDecoder
• Blastp & blastx against uniprot
• signalP
• HMMSCAN
• trinotate

Differential Expression Analysis

• align and estimate abundance (RSEM)
• abundance to matrix
• run DE analysis

GOSeq

• create files needed
• run GOSeq analysis

Genome-guided transcriptome assembly

snakemake -s refguided[SE/PE] -j <N_OF_JOBS> --use-singularity --singularity-args "-B <DATA_DIRECTORY>"

Sort alignment

• RSEM
• STAR
• samtools

Guided assembly

• trinity - assembly
• busco - checking quality
• cdhit - remove redundance

Annotation (Trinotate pipeline)

• TransDecoder
• Blastp & blastx against uniprot
• signalP
• HMMSCAN
• trinotate

Differential Expression Analysis

• align and estimate abundance (RSEM)
• abundance to matrix
• run DE analysis

GOSeq

• create files needed
• run GOSeq analysis