Metaerg
0. Introduction
Metaerg is an automated pipeline that uses third-party software as well as a large database to annotate genomes or sets of bins from microbial ecosystems. Examples of such annotation tasks are feature prediction with HMM, BLAST, and DIAMOND.
Learn more about Metaerg by reading their GitHub repository and the following article in Frontiers which was written by the same people who developed Metaerg.
1. Getting Started
1.1 Installation
1.2 Required Tools and Libraries
Perl Modules
| Dependencies |
|---|
| Archive::Extract |
| Bio::Perl |
| Bio::DB::EUtilities |
| DBD::SQLite |
| DBI |
| File::Copy::Recursive |
| HTML::Entities |
| LWP::Protocol::https |
| SWISS::Entry |
Table 1. This table displays all required Perl modules that are needed to run Metaerg
| Dependencies | Req. version | Version on Elja |
|---|---|---|
| antismash | ≥6.0.0 | 7.0.0 |
| ARAGORN | x | 1.2.41 |
| minced | x | 0.4.2 |
| BLAST+ executables | x | 2.13.0 |
| DIAMOND | 2.0.13 | 2.0.13 |
| GenomeTools | x | 1.6.2 |
| HMMER | 3.x.x | 3.3.2 |
| Infernal | x | 1.1.4 |
| prodigal | x | 2.6.3 |
| pyarrow | x | 12.0.0 |
| Python | x | 3.10.4 |
| RepeatMasker | x | 4.1.4 |
| RepeatScout | x | 1.0.6 |
| signalp | x | 0.5b |
| tmhmm | x | 2.0c |
| Tandem Repeats Finder | x | 4.09.1 |
Table 2. This table displays all main dependencies that are required to install Metaerg
2. Run Metaerg on Elja
2.1 Loading Metaerg
Before being able to run Metaerg on Elja, you will have to load the Metaerg module. To do this, type these commands in the terminal:
ml use /hpcapps/lib-mimir/modules/all
ml load Metaerg
2.2 Running Metaerg
To run Metaerg, you will have to type metaerg with parameters which you can find here. An example of a Metaerg run would look like this:
[..] $ metaerg --contig_file dir-with-contig-files --database_dir /AlphaFoldData/MetaergData/
Note that --database_dir /AlphaFoldData/MetaergData/ is always necessary as this is the location of the Metaerg database which the machine learning code is based on.