Metatranscriptomics-pipeline

This repository hosts a pipeline for analysing metatranscriptomic data. This is a comprehensive pipeline launched using Nextflow and incorporates a Singularity container. It performs de novo assembly and differential expression analysis allowing for better identification of novel genes or transcripts.

Running the pipeline

The pipeline is intended to be used on a cluster or server. The simple command for running the pipeline is as follows:

nextflow run trinity.nf -c config -profile cluster

Additional options for the command are:

-c [REQUIRED] path to the configuration file

-profile [REQUIRED] defines the infrastructure being used for execution, cluster for running on cluster and server for running on local servers

--reads [REQUIRED] full path to the 4 column tab delimited specimen file

--seqtype [REQUIRED] defines the type of reads, can either be or , default is fq

--libtype [REQUIRED] defines the strand-specific RNA-seq read orientation. For pair-ended reads it can either be FR or RF, for single reads it can either be F or R

--mem [OPTIONAL] maximum memory available for the algorthm to run. Input should be a string in the format "10G". Default value within the pipeline is 50G, this value will depend on available resources on execution infrastructure and the size of the dataset

--cpus [REQUIRED] number of cpus to be utilized at each stage of the assembly process. The default value is 8

--dispersion [REQUIRED] for differential expression analysis. Default is 0.1

--bfly_HeapSpaceMax [OPTIONAL] parameter for Butterlfy execution, default is 30G

--bfly_HeapSpaceInit [OPTIONAL] parameter for Butterfly execution, default is 5G

--bfly_GCThreads [OPTIONAL] specifies threads to be made availble for Butterfly execution, default is 10

--bfly_CPU [OPTIONAL] specifies maximum CPUs for Butterly algorithm, default is 6

--bowtie2_thr [OPTIONAL] defines the threads to be made available for bowtie2, default is 8

--outdir [OPTIONAL] specifies the name of the output directory to save pipeline output to. Default name is results.

Requirements for running pipeline

Nextflow 21 or higher

Singularity 3.5 or higher

Input

The pipeline processes raw paired-end or single-end fastq reads. The pipeline takes a 4 column tab-delimited file (for paired-end reads) and a 3 column tab-delimited file (for single-end reads). The tab-delimited specimen file specifies the sample ID, group name and absolute paths of raw reads.

Output

A directory named is saved in the current working directory. The following output is saved:

1. butterfly.txt - log file for last stage of assembly where the Butterfly algorithm does analysis.

2. chrysalis.txt - log file for assembly stage where Chrysalis algorithm does analysis.

3. DE - directory consisting of output from differential expression analysis. MA plots and Volcano plots comparing groups are saved here. The matrices used to generate the plots are saved here.

4. inchworm.txt - log file to record Inchworm algorithm during assembly.

5. jellyfish.txt - log file for the first step of assembly.

6. trinity_matrix - directory consisting of the count matrix.

7. trinity_out_dir - directory consisting of assembly intermediate files and trimmed reads.

8. trinity_out_dir.Trinity.fasta - assembly output with all assembled transcript in fasta format.

9. trinity_out_dir.Trinity.fasta.gene_trans_map - gene mapping file.

10. trinity_quantification - directory consisting of isoform and gene counts.

Software available in container

The container run Trinity (version 2.14.0). Development of the container and dependency software are specified in the definition file.