MetaErg

MetaErg is a stand-alone and fully automated metagenomic and metaproteomic data annotation pipeline. It bundles essential annotation tasks such as feature prediction, functional annotation wit h Hidden Markov Model (HMM) searches as well as blast and diamond searches. It estimates and visualizes quantitative taxonomic and pathway compositions of multiple metagenomic and proteomics s amples using sequencing coverage and proteomics spectral counts, respectively. For visualization, MetaErg provides a HTML interface, bringing all annotation results together in sortable and se archable tables, collapsible trees, and other graphic representations, enabling intuitive navigation of complex data.

A MetaErg analysis output demo page can be found at: https://xiaoli-dong.github.io/metaerg/

A MetaErg Docker application can be found here: https://hub.docker.com/r/xiaolidong/docker-metaerg

Please cite the following: Dong X and Strous M (2019) An Integrated Pipeline for Annotation and Visualization of Metagenomic Contigs. Front. Genet. 10:999. doi: 10.3389/fgene.2019.00999

Required perl modules

If you do not use Docker, you will need to install perl modules. MetaErg requires Perl 5.6.0 or higher and runs on Linux platforms. Besides the perl core modules, it also requires the followin g perl modules to be installed:

* Archive::Extract;
* Bio::Perl;
* Bio::DB::EUtilities
* DBD::SQLite
* DBI;
* File::Copy::Recursive
* HTML::Entities
* LWP::Protocol::https
* SWISS::Entry;
* SWISS::KW;

Third-party software

If you do not use Docker, you will need to install the following 3rd party dependencies and make sure they are on your system path:

• ARAGORN: a program to detect tRNA genes and tmRNA genes in nucleotide sequences
• BLAST+ executables: The Basic Local Alignment Search Tool (BLAST) finds regions of local similarity between sequences.
• DIAMOND: a new high-throughput program for aligning DNA reads or protein sequences against a protein reference database
• Hmmer: HMMER is used for searching sequence databases for sequence homologs, and for making sequence alignments.
• MinCED: a program to find Clustered Regularly Interspaced Short Palindromic Repeats (CRISPRs) in full genomes or environmental datasets such as asse mbled contigs from metagenomes.
• MinPath: a parsimony approach for biological pathway reconstructions using protein family predictions, achieving a more conservative, yet more faithful, estimation of the biological pathways for a query dataset.
• Prodigal: Fast, reliable protein-coding gene prediction for prokaryotic genomes.
• SignalP: The program predicts the presence of signal peptides and the location of their cleavage sites in proteins from Archaea, Gram- positive Bacteria, Gram-negative Bacteria and Eukarya.
• TMHMM: a method for predicting transmembrane helices based on a hidden Markov model

MetaErg reference DB

MetaErg requires external databases that need to be downloaded and unarchived

# Retrieve the prebuilt database (The current database were generated on Jan. 8, 2021)
wget http://ebg.ucalgary.ca/metaerg/db.tar.gz -P $HOME
tar -xvzf $HOME/db.tar.gz

or built using the supplied script (see running with docker and installation sections).

MetaErg databases were built based on the following publicly available databases:

• casgene.hmm
• FOAM
• metabolic-hmms
• Pfam
• SwissProt
• SILVA
• TIGRFAMS
• GTDBTK
• RefSeq

Running with docker

The MetaErg docker image is hosted on the docker hub: https://hub.docker.com/r/xiaolidong/docker-metaerg. Due to licencing permissions, this image does not contain SignalP and TMHMM. When running with docker image, "--sp --tm" options cannot be enabled.

# Get the Docker image
docker pull xiaolidong/docker-metaerg

# Databases and Docker
With Docker, you can either use the downloaded prebuilt database or build a database with the command below. Building the database process will take a while to run:
docker run --shm-size 2g --rm -u $(id -u):$(id -g) -it -v my_local_dir:/data/ xiaolidong/docker-metaerg setup_db.pl -o /data -v 132

#Running MetaErg with default options
docker run --shm-size 2g --rm -u $(id -u):$(id -g) -it -v my_local_dir:/data/ xiaolidong/docker-metaerg metaerg.pl --dbdir /data/db --outdir /data/my_metaerg_output /data/contig.fasta

Installation

# Install metaerg to your home directory
git clone https://github.com/xiaoli-dong/metaerg.git $HOME/metaerg

# Using the downloaded prebuilt database
# or build database using MetaErg supplied script and the building process will take a while to run
$HOME/metaerg/bin/setup_db.pl -o $HOME -v 132

Quick start:

The functionality provided by MetaErg can be accessed through the help menu:

>perl $HOME/metaerg/bin/metaerg.pl --help

Running MetaErg with the default parameters will output the final and intermediate results into a directory named metaerg.pl_ddmmyyyy. Without --dbdir option, MetaErg will look for the databas e directory "db" inside the metaerg directory

>perl $HOME/metaerg/bin/metaerg.pl --dbdir $home/db contig.fasta

Running MetaErg with user defined output directory and file names:

>perl $HOME/metaerg/bin/metaerg.pl --dbdir $home/db --outdir mydir --prefix mycontigs contig.fasta

With a user provided "depth file", MetaErg can quantify the taxonomic, functional, and pathway compositions of multiple metagenomic samples. An example "depth file" is included in the "example " directory. The depth file was generated with the script "jgi_summarize_bam_contig_depths" from MetaBat using BAM files. BAM files are created by aligning the reads of each metagenomic sample to the assembled contigs, using a program such as BBMap or bwa.

>perl $HOME/metaerg/bin/metaerg.pl --dbdir $home/db --depth demo.depth.txt demo.fasta

Utility scripts

MetaErg includes some utility perl scripts and they can be used to filter contigs by length, add bin identifiers to predicted coding sequences:

#Filter out contig sequences shorter than 500bp
>perl $HOME/metaerg/bin/filterContigByLength.pl contig.fasta 500

Let's assume you have annotated all the contigs of your metagenome. MetaErg can extract the subset of annotation results and produce html summary pages for an individual bin as follows:

#Extract annotation results for individual bins
#Step1, extracting the gff-format annotations for the contigs included in "mybin.nt.fasta" from the total metaerg dataset annotation:
>perl $HOME/metaerg/bin/fastaContig2Gff.pl -c mybin.nt.fasta -g mydir/data/all.gff  > mybin.gff

#Step 2, generating the html reports for the extracted contig subset
>perl $HOME/metaerg/bin/output_reports.pl  -g mybin.gff -f subset.fasta -o mybindir

Let's assume mybindir contains many nucleotide fasta files, one for each bin: Bin.1.fa", "Bin.2.fa", "Bin.3.fa"... files. The following commands will:

#Add bin id to the fasta format of the protein coding sequence and protein coding sequence id will be in the format of "binid_geneid"
>perl $HOME/metaerg/bin/add_binid2cds.pl -d binning -c mydir/data/cds.faa -g mydir/data/all.gff

#Add bin ids to master.tsv file, as the first column
>perl $HOME/metaerg/bin/add_binid2master_dot_tsv.pl -d binning -t mydir/data/master.tsv

MetaErg output directory layout

MetaErg writes the output files into a user defined or MetaErg autogenerated output directory. An example of the MetaErg output directory layout is as following:

xact same parameters, MetaErg will start from the place it failed. After a MetaErg job finishes successfully, this directory can be deleted before you transfer the results to your local compu ters|

MetaErg data directory file descriptions