Transcriptome Mapping Pipeline

https://github.com/palfalvi/rnaseq

This pipeline is a basic workflow to QC and quantify raw RNA-seq reads with the following steps:

1, Raw fastq quality check and filtering

Utilizing fastp software, quality values are recorded and low quality reads removed. fastp also recognizes and trims adaptors.

2, Index generation

This depends on the quantification mode selected. Currently salmon, kallisto and star are supported.

3, Mapping and quantification

This is connected to the index generation. 'salmon' and 'kallisto' natively quantifies reads, while the 'star' mode utilizes featureCounts

Dependencies

As this is a nextflow project dependent on conda environments, the only 2 things you need before running are nextflow and conda:

Install nextflow

In your bin directory, or in some places you can access from your $PATH, copy and execute the following:

wget -qO- https://get.nextflow.io | bash

Done.

Install miniconda

Please download the miniconda installer into your bin directory form the miniconda repository and execute the follwing :

bash Miniconda3-latest-Linux-x86_64.sh

Done.

• If it did not work on NIBB-BIAS5, please try on bias5-db.

SRA accessions [currently not supported. Coming soon.]

If you wish to use SRA ids directly, you can provide with --sra SRP043510 instead of --reads. This feature, however uses the NCBI Esearch API, which requires an NCBI API Key in your environment. To get an API Key, follow this link.

After you get your API key, you need to set it into your environment. In your favorite editor (e.g. nano or vim) open ~/.bashrc and in the end of the file, insert the following: export NCBI_API_KEY=0123456789abcdef then save. You need to run source ~/.bashrc or log out and in again to make it work. From that point, you do not need to modify or rerun this part.

How to run

First, you need to decide which mode you would like to run. The default and recommended way is salmon, but kallisto and star are also available. Just change specify with the --mode flag. e.g --mode star.

Then, you need to know if you have pair-end or single-end dataset. The pipeline runs by default on pair end data, but you can specify single end mode with the --single flag.

If you run in single end mode with kallisto, please also specify --fragment_length and --fragment_sd, which can be calculated from your BioAnalyzer file of the final libraries.

Finally, you need to have a reference set. For salmon and kallisto, this should be a fasta file of transcripts (not genes!) specified with the --transcriptome flag. In the case of star, please specify the genome and a corresponding GTF annotation file with --genome and --gtf flags.

Example commands

Salmon mapping with pair end reads.

nextflow run palfalvi/rnaseq --reads /path/to/reads/*R{1,2}.fastq --transcriptome transcripts.fasta

Kallisto with single end reads.

nextflow run palfalvi/rnaseq --mode kallisto --reads /path/to/reads/*.fastq --transcriptome transcripts.fasta --fragment_length 300 --fragment_sd 30

STAR with pair end reads

nextflow run palfalvi/rnaseq --mode star --reads /path/to/reads/*R{1,2}.fastq --genome genome.fastq --gtf annotation.gtf

Salmon with no QC and trimming and save the index files for later use. Also, save the results into salmon_output directory.

nextflow run palfalvi/rnaseq --reads /path/to/reads/*R{1,2}.fastq --transcriptome transcripts.fasta --save_index --skip_qc --out salmon_output