Melange: A Snakemake workflow that streamlines structural and functional annotation of prokaryote genomes

Full documentation: https://sandragodinhosilva.github.io/melange/

1 Overview

• Melange is a genome annotation tool that enables the simultaneous annotation of large genome datasets using multiple databases: Pfam, COG, KEGG, CAZymes and/or MEROPS.
• Melange can handle unassembled and assembled sequencing data and amino acid sequences, with automatic download and configuration of necessary tools and databases.
• As a Snakemake pipeline, Melange is highly scalable, reproducible and has a transparent workflow, and can be used to annotate one to thousands of genomes, producing several easy-to-analyze tabular outputs.
• Melange derives its name from the French term "mélange", which signifies a blend or collection of diverse entities. This choice of name is a tribute to its ability to facilitate annotation using multiple databases.

Sinopse: Melange - a versatile and user-friendly genome annotation tool that enables the simultaneous annotation of large genome datasets using multiple databases. Melange implementation in Snakemake allows flexibility and scalability. The unified output tables facilitate further analysis and are suitable for various comparative studies.

2 System requirements

• Melange is designed to run on Linux systems and requires installation of Python (v≥3.8), Snakemake (v≥5.19.2) and Conda (v≥4.10.1). Optionally, git can also be installed for easy download of the repository.
• Only databases selected in the config.yml file are downloaded and configured locally, reducing storage requirements to run Melange.
• A test dataset (available in the "example_data" directory) is provided to allow for a test run to confirm the correct installation.
• Melange utilizes Snakemake for modularity and automatic parallelization of jobs, making it suitable for implementation on high-performance computational clusters. For more details on that check "Run on HPCs".

3 Melange features

Melange allows the simultaneous functional annotation of prokaryote genomes or metagenomes with multiple annotation schemes. Here are illustrated all Melange features:

• A) Input files: unassembled fastq, nucleotide fasta or amino acid fasta files (to be configured in config.yml)
• B) Genome annotation: annotation databases (and respective search algorithm used to query the proteins): Pfam, COG, KEGG, CAZymes or MEROPS (to be configured in config.yml)
• C) Outputs

A) Inputs

Melange accepts 3 types of input files:

• unassembled (meta)genomic data (.fastq)
• (meta)genome assemblies (.fna, .fasta, .ffn, .faa, .frn, .fa)
• predicted amino acid sequences (.faa)

If fastq files are inputted, Melange will convert them to fasta nucleotide files using the EMBOSS tool seqret before annotation.

B) Genome annotation

B1) Gene calling and general annotation

When nucleotide files are submitted, Melange first performs a structural annotation step using Prokka v1.14.5 [1] with default settings and outputs the corresponding translations into amino acid sequences in a fasta file. In addition to this output, which will be used in all subsequent steps, Prokka also generates other additional file formats, such as GenBank files, per genome.

B2) Functional annotation

Melange allows functional annotation of genomes with up to five databases: Pfam, COG, KEGG, CAZymes and MEROPS.

Here are described the main characteristics of the annotation procedure with each database:

Pfam: For the annotation with Pfam identifiers, a local database is created using HMMER v3.3 from the latest version of Pfam-A.hmm file (currently v35.0) downloaded from the downloaded from the InterPro repository A local database is constructed using HMMER v3.3. Once the local database has been created, query proteins are searched against it using the hmmscan function from the HMMER suite. The best hit per ORF (cut-off: -E 1e-5) is selected. Selected references: [2,3].

COG (Clusters of Orthologous Groups): The COG annotation procedure follows the cdd2cog v0.2 workflow. First, several files are downloaded from the NCBI's FTP server, including a preformatted database of the NCBI's Conserved Domain Database (CDD) COG distribution (2020 release). Query proteins are then blasted against this database using reverse position-specific BLAST (rps-blast) function from the Blast+ v2.9.0 suite and the results are parsed to a readable format with a Perl script (cdd2cog.pl). The best hit per ORF (cut-off: -E 1e-5) is selected. Selected references: [4].

KEGG (Kyoto Encyclopaedia of Genes and Genomes): To obtain the KEGG Orthology (KO) for protein identification, the command line (CLI) version of KofamKoala(https://www.genome.jp/tools/kofamkoala/) - Kofamscan - is used. Kofamscan performs K number assignments using hidden Markov model (HMM) profile search, which involves searching query proteins against a customized HMM database of KOs (KEGG release 103.0). This database includes predefined thresholds for individual KOs, resulting in more reliable assignments than sequence similarity searches. Kofamscan uses the hmmsearch function from the HMMER suite to perform the search. Selected references: [5,6].

CAZymes (Carbohydrate-active enzymes): The CAZymes annotation procedure uses the meta server dbCAN2, specifically, the standalone version run_dbcan v2.0.11 implemented with default settings. Run_dbcan is a tool that performs annotation of CAZymes using three different approaches: a HMMER v3.3 search against the dbCAN HMM database, a DIAMOND v0.9.32 search against the CAZy database, and the eCAMI algorithm. For improved annotation accuracy, ORFs are only annotated with the respective CAZyme name if at least two database searches were positive, as suggested by dbCAN2 authors in Zhang et al.. Selected references: [7,8].

MEROPS: For the identification of ORFs encoding for peptidases and their inhibitors the "merops_scan.lib", release 12.4 file is downloaded from MEROPS. Then makedblast is used to produce a local BLAST database. Query aminoacid sequences are then searched for matches with this database with blastp. Selected references: [9].

C) Outputs

Melange produces several different output formats tailored to meet users' diverse needs, with almost no additional computational cost. This is achieved by leveraging the output of each annotation database and transforming it into different tables.

In summary, three files with distinct data representation modes are created for each annotation type:

• Counts
• Presence/absence (PA)
• Relative abundance

In these output tables, each row represents a database identifier (ID), and each column represents an input (either nucleotide or amino acid (meta)genome files). While in counts (A), nij represents the number of proteins or protein domains (depending on the database in use) identified with a certain ID for a given input, in the PA table (B), nij equal to 1 indicates the existence of a certain identifier in the input, and 0 indicates its absence. In the relative abundance annotation table (C), nij represents the normalized count of an ID per the total number of ORFs in each input.

In addition to the annotation tables, Melange also provides:

• intermediate files - including different file types (e.g. GenBank, GFFF, etc);
• descriptive file (e.g. Pfam_description.csv) containing a summarized description of each annotation ID;
• statistics.csv - % of Orfs annotated with each database;
• folder Orf_per_genome: each genome has a unique file containing all orfs identified by Prokka and the subsequent annotations with the selected functional databases;
• benchmark results - Melange automatically records running metrics using the Snakemake directive benchmark.

4 Usage

This is a simple description on how to use melange. For more details, please see Melange documentation.

Step 0: Install conda and Snakemake

Conda and Snakemake are required to be able to use Melange.
Conda is easy to install via its lightweight version Miniconda.
After installing Conda, install Snakemake:

# As described in Snakemake documentation:
conda install -c conda-forge mamba
mamba create -c conda-forge -c bioconda -n snakemake snakemake
conda activate snakemake

Step 1: Clone workflow

To use Melange, you need a local copy of the workflow repository. Start by creating a clone of the repository:

git clone https://github.com/sandragodinhosilva/melange.git

Step 2: Configure workflow

Configure the workflow according to your needs by editing the file config.yaml.

Here you can select which databases (Pfam, COG, Kegg, CAZymes and/or MEROPS) are to be used.

You can also define if input files are either fasta nucleotide files (e.g. fna, fa) or fasta aminoacid files.

More information about configuration settings can be found at: config/README.md

Step 3: Execute workflow

Execute the workflow locally with N cores:

snakemake --use-conda --cores N

Execution on a cluster, example:

snakemake --use-conda --cluster qsub --jobs 8

For more information about running on a computational cluster, please check snakemake documentation about it: https://snakemake.readthedocs.io/en/stable/executing/cluster.html

---

Citing Melange

At the moment, Melange does not have a publication describing its features (we are working on it). Please use a link to Melange Github when referring to this tool.

Melange Contributions

• Sandra Godinho Silva ^1,2 - MicroEcoEvo - iBB, IST.
• Tina Keller-Costa ^1,2 - MicroEcoEvo - iBB, IST.
• Sanchita Kamath ³ - Microbial Systems Data Science - UFZ, Leipzig.
• Ulisses Nunes da Rocha ³ - Microbial Systems Data Science - UFZ, Leipzig.
• Rodrigo Costa ^1,2 - MicroEcoEvo - iBB, IST.

¹ Institute for Bioengineering and Biosciences, Department of Bioengineering, Instituto Superior Técnico da Universidade de Lisboa, Lisbon, Portugal
² Associate Laboratory, Institute for Health and Bioeconomy, Instituto Superior Técnico, University of Lisbon, Lisbon, Portugal
³ Department of Environmental Microbiology, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany

Funding

We acknowledge national funds from the Portuguese Foundation for Science and Technology (FCT), in the scope of the projects UIDB/04565/2020 and UIDP/04565/2020 of iBB and the project LA/P/0140/2020 of i4HB. Computational support was received from INCD funded by FCT and FEDER under project 01/SAICT/2016 n° 022153. SGS is the recipient of a PhD scholarship conceded by the FCT (PD/BD/143029/2018) and was supported by a research and training grant conceded by the Federation of European Microbiological Societies (FEMS-GO-2019-511). TKC acknowlesdges an investigator contract (CEECIND/00788/2017) conceded by the FCT. UNR was financed by the Helmholtz Association (VH-NG-1248 Micro' Big Data') and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – project number 460129525.

---

Selected References

• [1] Seemann T. Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30(14):2068-9.
• [2] El-Gebali S, Mistry J, Bateman A, Eddy SR, Luciani A, Potter SC, et al. The Pfam protein families database in 2019. Nucleic Acids Research. 2019;47(D1):D427-D32.
• [3] Blum M, Chang H-Y, Chuguransky S, Grego T, Kandasaamy S, Mitchell A, et al. The InterPro protein families and domains database: 20 years on. Nucleic Acids Research. 2021;49(D1):D344-D54.
• [4] Galperin MY, Wolf YI, Makarova KS, Vera Alvarez R, Landsman D, Koonin EV. COG database update: focus on microbial diversity, model organisms, and widespread pathogens. Nucleic Acids Research. 2021;49(D1):D274-D81.
• [5] Kanehisa M, Furumichi M, Sato Y, Ishiguro-Watanabe M, Tanabe M. KEGG: integrating viruses and cellular organisms. Nucleic Acids Research. 2021;49(D1):D545-D51.
• [6] Ogata H, Goto S, Kanehisa M, Ohkubo K, Endo H, Blanc-Mathieu R, et al. KofamKOALA: KEGG Ortholog assignment based on profile HMM and adaptive score threshold. Bioinformatics. 2020;36(7):2251-2.
• [7] Lombard V, Golaconda Ramulu H, Drula E, Coutinho PM, Henrissat B. The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Research. 2014;42(Database issue):D490-5.
• [8] Zhang H, Yohe T, Huang L, Entwistle S, Wu P, Yang Z, et al. dbCAN2: a meta server for automated carbohydrate-active enzyme annotation. Nucleic Acids Research. 2018;46(W1):W95-W101.
• [9] Rawlings ND, Barrett AJ, Thomas PD, Huang X, Bateman A, Finn RD. The MEROPS database of proteolytic enzymes, their substrates and inhibitors in 2017 and a comparison with peptidases in the PANTHER database. Nucleic Acids Research. 2018;46(D1):D624-D32.