PLATE: Plasmid Atlas of Food Ecosystems

If you use this repository, please cite:

Aslan MU, Kayacan RA, Söğütdelen SA, Sağıroğlu Ö, Arıkan M. PLATE: a global atlas of food metaplasmidomes, bioRxiv, 2026 [Link].

Overview

PLATE is a comprehensive resource of food metaplasmidomes, built with an end-to-end pipeline that retrieves metagenomic assemblies, identifies putative plasmid sequences and annotates them.

The PLATE datasets are publicly available through the links below:

Catalog	Catalog link
bPLATE - Sequences	Link
bPLATE - Metadata	Link
rPLATE - Sequences	Link
rPLATE - Metadata	Link

PLATE is also accessible as a searchable catalog at https://plate-db.com/

Step-by-Step Workflow

PLATE was constructed by:

Retrieving metagenomic assemblies (or assembling them),
Predicting plasmid contigs with Plasmer,
Removing identical contigs with seqkit and grouping the unique contigs into plasmid-like clusters with MMseqs2,
Predicting genes and smORFs on plasmid contigs,
Assigning hosts to plasmid contigs,
Running functional annotation tools (protein families, resistance genes, CAZymes, CRISPR-Cas, AMPs and ACPs),
Constructing a refined plasmid set by passing the Plasmer-predicted plasmid contigs through five successive filters (PlasClass, viralVerify, PlasFlow, Infernal and minimap2).

Part 1 — Data acquisition

LOGAN

Metagenomic contigs are retrieved from the LoganSRA repository. The assemblies are compressed and must be downloaded and extracted.

Usage

wget https://s3.amazonaws.com/logan-pub/c/$accession/$accession.contigs.fa.zst
zstd -d $accession.contigs.fa.zst

Notes

accession: The SRA run accession ID (e.g., ERR123456).
zstd -d: Decompresses the .zst file into a standard FASTA file (.fa).

Assembly (optional)

Required only if your data is not available via LOGAN. Unitigs are first built with Cuttlefish2, then assembled into contigs with Minia3.

Cuttlefish2

Setup

conda create -n cuttlefish -c bioconda -c conda-forge -c defaults python=3.10 sra-tools seqkit "cmake>=3.16" make gxx_linux-64 zlib git -y
conda activate cuttlefish

mkdir -p vendor
cd vendor
git clone https://github.com/rchikhi/cuttlefish.git
cd cuttlefish
git checkout 9401ef5

mkdir -p build
cd build
export CMAKE_POLICY_VERSION_MINIMUM=3.5
cmake -DCMAKE_INSTALL_PREFIX=../ \
      -DPYTHON_EXECUTABLE="$(which python3)" \
      ..
make -j "$(nproc)" install

mkdir -p bin
ln -sf vendor/cuttlefish/bin/cuttlefish bin/cuttlefish

Usage

cuttlefish build --ref -s input.fa -k 31 -c 2 -t 8 -o output.unitig.fa

Notes

--ref : Builds a compacted reference de Bruijn graph (for FASTA input)
-s : Input FASTA file
-k : k-mer length
-c : Minimum k-mer abundance cutoff
-t : Number of threads
-o : Output file (unitigs)

Minia3

Setup

conda create -n minia3 python=3.10 -y
conda activate minia3

git clone https://github.com/GATB/minia.git
cd minia
git checkout 71484e8
git submodule update --init --recursive

mkdir build && cd build
CMAKE_POLICY_VERSION_MINIMUM=3.5 cmake ..
make -j$(nproc)

Usage

minia \
  -in input.unitig.fa \
  -kmer-size 31 \
  -abundance-min 1 \
  -traversal contig \
  -out output

Notes

-in : Input file (unitigs from Cuttlefish2)
-kmer-size : k-mer length used for assembly
-abundance-min : Minimum k-mer abundance threshold
-traversal : Graph traversal mode (builds contigs from the unitig graph)
-out : Output prefix

Part 2 — Plasmid prediction

Plasmer is run on the metagenomic contigs from Part 1. Its predicted plasmid contigs (predPlasmid.fa) are the input of Parts 3–7.

Plasmer

Setup

conda create -n plasmer -c bioconda -c conda-forge
conda activate plasmer

conda install -y seqkit=2.2.0 python=3.10.4 biopython prodigal=2.6.3 hmmer=3.3.2 "blast>=2.10.1" infernal=1.1.4 diamond=2.0.8 parallel=20220722 kraken2=2.1.2 r-base=4.2.0 r-randomforest
conda install -y -c bioconda kmer-db=1.9.4

git clone https://github.com/nekokoe/Plasmer.git
cd Plasmer
export PATH=$PATH:$(pwd)

Install the databases

The pre-built database are available at Zenodo and Google Drive.

Verify the downloaded databases:

$ sha1sum plasmerMainDB.tar.xz 
0b08f5c30d60b137f54de6024ab7557031850db6  plasmerMainDB.tar.xz

$ sha1sum customizedKraken2DB.tar.xz 
b14efdd9232fd5f6d066716bd8e3e6ca80c9c0de  customizedKraken2DB.tar.xz

Extract the databases:

mkdir -p db_Plasmer
tar -xJvf plasmerMainDB.tar.xz -C db_Plasmer
tar -xJvf customizedKraken2DB.tar.xz -C db_Plasmer

Usage

./Plasmer/scripts/Plasmer -g input.fa -p $accession -d db_Plasmer -t 3 -m 500 -l 500000 -o outpath

Notes

-g : Input FASTA file (contigs)
-p : Output file prefix
-d : Path to the Plasmer database
-t : Number of CPU threads
-m : Minimum contig length (bp)
-l : Maximum contig length (bp)
-o : Output directory. Predictions are saved as .predClass.tsv and separate FASTA files.

Part 3 — Dereplication and clustering

Identical contigs are first collapsed with seqkit, then the unique contigs are grouped into plasmid-like clusters with MMseqs2.

seqkit

Removes identical contigs from the Plasmer-predicted plasmid contigs and from the refined plasmid set (Part 7). Contigs with the same sequence on either strand (reverse complements count as identical) are collapsed into one; the first record is kept.

Setup

conda create -n seqkit -c conda-forge -c bioconda seqkit
conda activate seqkit

Usage

seqkit rmdup -s -j 4 -D duplicates.txt input.fa > output.unique.fa

Notes

rmdup : Removes duplicated sequences
-s : Compares sequences instead of names; both strands are compared
-j : Number of threads
-D : Output file of the duplicated contigs (number of copies and their names)
output.unique.fa : Output FASTA file with one copy of every sequence

MMseqs2

Clusters the unique contigs of the refined plasmid set (seqkit output) into plasmid-like clusters. A contig joins a cluster when it is ≥ 95% identical to the cluster representative and the alignment covers ≥ 85% of the contig; the longest contig of each cluster is its representative.

Setup

conda create -n mmseqs2 -c conda-forge -c bioconda mmseqs2
conda activate mmseqs2

Usage

mmseqs easy-cluster input.fa output_prefix tmp_dir \
    --min-seq-id 0.95 \
    -c 0.85 \
    --cov-mode 1 \
    --cluster-mode 2 \
    --alignment-mode 3 \
    --kmer-per-seq-scale 0.2 \
    --max-seq-len 500000 \
    --threads 4 \
    --split-memory-limit 9G

Notes

output_prefix : Output prefix (_cluster.tsv, _rep_seq.fasta, _all_seqs.fasta)
tmp_dir : Temporary directory
--min-seq-id : Minimum sequence identity to the cluster representative
-c : Minimum alignment coverage
--cov-mode : Coverage is measured on the cluster member
--cluster-mode : Greedy incremental clustering by length (the longest contig represents its cluster)
--alignment-mode : Computes the alignment score, coverage and sequence identity
--kmer-per-seq-scale : Number of k-mers per sequence, scaled with sequence length
--max-seq-len : Maximum sequence length (the default, 65,535 bp, is shorter than the longest contigs)
--threads : Number of threads
--split-memory-limit : Maximum memory used per database split

Part 4 — Gene prediction

Prodigal and smORFinder are run in meta mode on the predicted plasmid contigs (predPlasmid.fa, the output of Plasmer).

Prodigal

Setup

git clone https://github.com/hyattpd/Prodigal.git
cd Prodigal/
make install

Usage

prodigal -p meta \
         -i input.fa \
         -o output.gff \
         -a output.faa \
         -d output.fna \
         -f gff

Notes

-p : Mode (single/meta)
-i : Input FASTA file
-o : Gene coordinates (.gff)
-a : Predicted protein sequences (.faa)
-d : Predicted gene nucleotide sequences (.fna)
-f : Output format

smORFinder

Setup

conda create -n smorfinder python=3.8.20
conda activate smorfinder

pip install smorfinder

Usage

smorf meta input.fa -o outpath -t $threads

Notes

meta : Runs smORFinder in metagenomic mode
input.fa : Input FASTA file (plasmid contigs)
-o : Output directory
-t : Number of threads

Part 5 — Host assignment

HOTSPOT is run on the predicted plasmid contigs (predPlasmid.fa).

HOTSPOT

Setup

git clone https://github.com/Orin-beep/HOTSPOT
cd HOTSPOT/
conda env create -f environment.yaml -n hotspot
conda activate hotspot

Install the databases:

sh prepare_db.sh        # download and unzip the database folder
sh prepare_mdl.sh       # download and unzip the model folder

Usage

python preprocessing.py \
        --fasta input.fa \
        --database folder/of/database \
        --model_path folder/of/model \
        --threads $threads

python hotspot.py --threads $threads --out outpath --device cpu

Notes

preprocessing.py : Prepares the input sequences for prediction
--fasta : Input FASTA file (plasmid contigs)
--database : Path to the downloaded database folder (PC proteins, MOB/MPF proteins and replicons)
--model_path : Path to the downloaded model folder
--threads : Number of threads
hotspot.py : Predicts the hosts of the preprocessed sequences
--out : Output directory for the prediction results
--device : Device used for prediction ('gpu' or 'cpu')

Part 6 — Functional annotation

Protein-level tools use the Prodigal proteins (.faa), Macrel and MetaPepticon use the smORFinder peptides, and dbCAN (CGC mode) and CRISPR-Cas Typer use the plasmid contigs directly.

eggNOG-mapper

Identifies protein family profiles of plasmid contigs.

Setup

conda create -n emapper -c bioconda -c conda-forge python=3.10.19
conda activate emapper
conda install eggnog-mapper

Install the databases:

aria2c -c -x 8 -s 8 -k 1M http://eggnog5.embl.de/download/emapperdb-5.0.2/eggnog.db.gz
aria2c -c -x 8 -s 8 -k 1M http://eggnog5.embl.de/download/emapperdb-5.0.2/eggnog_proteins.dmnd.gz
aria2c -c -x 8 -s 8 -k 1M http://eggnog5.embl.de/download/emapperdb-5.0.2/eggnog.taxa.tar.gz

Usage

emapper.py \
    -i input.faa \
    -o output_prefix \
    --output_dir output_dir \
    --data_dir eggnog_db_dir \
    --scratch_dir tmp_dir \
    --cpu 8 \
    -m diamond \
    --override \
    --dmnd_ignore_warnings

Notes

-i : Input protein FASTA file
-o : Output file prefix
--output_dir : Output directory
--data_dir : Path to EggNOG database files (eggnog.db, eggnog_proteins.dmnd, eggnog.taxa.db)
--scratch_dir : Temporary directory
--cpu : Number of CPU threads
-m diamond : Uses Diamond for homology search
--override : Overwrites existing output files
--dmnd_ignore_warnings : Tolerates ambiguous amino acids (e.g., X)

AMRFinderPlus

Detects resistance genes (antimicrobial, stress, heat, heavy-metal, etc.).

Setup

conda create -y -c conda-forge -c bioconda -n amrfinder --strict-channel-priority ncbi-amrfinderplus
conda activate amrfinder

Usage

amrfinder -p input.faa --plus --threads $threads -o output.tsv

Notes

-p : Protein FASTA file
--plus : Adds "Plus" genes (virulence factors, stress-response genes, etc.)
-o : Output file path
--threads : Number of threads

Resistance Gene Identifier

Detects antimicrobial resistance (AMR) genes using the CARD database.

Setup

conda create --name rgi -c conda-forge -c bioconda -c defaults 
conda activate rgi
conda install -c conda-forge -c bioconda -c defaults rgi

CARD database setup (system-wide):

rgi load --card_json /path/to/card.json

CARD database setup (local/working directory):

rgi load --card_json /path/to/card.json --local

Usage

rgi main -a DIAMOND \
         -i input.faa \
         -o outpath \
         --local \
         --clean \
         -t protein \
         -n $threads

Notes

load : Loads the CARD database
--card_json : Path to the CARD database JSON file
main : Runs AMR gene detection
-a : Alignment tool
-i : Input file path
-o : Output file path
--local : Uses the local database
--clean : Removes temporary files
-t : Input type (contig or protein)
-n : Number of threads

dbCAN

Identifies CAZymes and CAZyme gene clusters.

Setup

git clone https://github.com/bcb-unl/run_dbcan.git
cd run_dbcan
conda create -f envs/environment.yml
conda activate run_dbcan_env
pip install dbcan

Install the databases:

mkdir -p db
run_dbcan database --db_dir db

Usage

# for CAZyme Annotations
run_dbcan CAZyme_annotation \       
    --mode protein \
    --input_raw_data input.faa \
    --db_dir db \
    --output_dir outpath

# for CAZyme Gene Clusters
run_dbcan easy_CGC \                
   --mode meta \
   --input_raw_data input.fasta \
   --db_dir db \
   --output_dir outpath

Notes

--mode : Input type. protein for CAZyme_annotation, meta (plasmid contigs) for easy_CGC
--input_raw_data : Input file (.faa or contig FASTA)
--db_dir : Database directory
--output_dir : Output directory

CRISPR-Cas Typer

Detects CRISPR-Cas genes and arrays.

Setup

conda create -n cctyper -c conda-forge -c bioconda -c russel88 cctyper
conda activate cctyper
pip install "setuptools<70"

Usage

cctyper input.fasta output_dir --prodigal meta

Notes

--prodigal : Prodigal mode used by CCTyper (single/meta)

Macrel

Identifies antimicrobial peptides (AMPs).

Setup

conda create -n macrel python=3.10 -y
conda activate macrel
conda install -y -c bioconda -c conda-forge megahit paladin ngless
pip install macrel

Usage

macrel peptides \
  --fasta input.faa \
  --output output_dir \
  --threads 8

Notes

peptides : Predicts AMPs directly from peptide sequences
--fasta : Amino acid FASTA file (output of smORFinder)
--output : Output directory
--threads : Number of threads

MetaPepticon

Identifies anticancer peptides (ACPs).

Setup

conda create -c conda-forge -c bioconda -n snakemake snakemake=8.25.5 python=3.11
git clone --recursive https://github.com/arikanlab/MetaPepticon.git

Usage

bash workflow/scripts/prepare_config.sh
conda activate snakemake
snakemake -s workflow/Snakefile --resources slot=1 --cores 16 --use-conda

Notes

--resources slot=1 : Allows only one job that requires the slot resource (assembly and toxicity prediction steps) to run at a time
--cores : Number of CPU cores
--use-conda : Runs each rule in its own conda environment

Part 7 — Constructing the refined plasmid set

The refined plasmid set is built from the Plasmer-predicted plasmid contigs (predPlasmid.fa, Part 2) that are at least 2 kb long. The contigs pass through five successive filters; each filter takes the contigs kept by the previous one, and the contigs that remain after Filter 5 form the refined plasmid set.

Filter 1 — Plasmid-likeness scoring (PlasClass)

Scores each contig for plasmid-likeness from its k-mer composition (k = 3–7), using the pretrained models shipped with PlasClass. Contigs with a plasmid score ≥ 0.5 are kept.

Setup

git clone https://github.com/Shamir-Lab/PlasClass.git
cd PlasClass

conda create -y -p ./classification-env python=3.8
conda activate ./classification-env

pip install numpy==1.22.0 scipy==1.10.0 joblib==1.2.0 "cython<3"
pip install "setuptools<60" wheel
pip install --no-build-isolation scikit-learn==0.21.3

python setup.py install

Usage

python PlasClass/classify_fasta.py -f input.fa -o output.probs.out -p 8

Notes

-f : Input FASTA file (Plasmer-predicted plasmid contigs ≥ 2 kb)
-o : Output file (tab-separated contig name and plasmid score)
-p : Number of parallel processes

Filter 2 — Gene-content classification (viralVerify)

Classifies contigs as plasmid, chromosome or virus from their gene content (Prodigal gene prediction, hmmsearch against informative HMMs and a Naive Bayes classifier). Contigs classified as Virus or Chromosome are removed; Plasmid and all Uncertain classes are kept.

Setup

conda create -n viralverify -c conda-forge -c bioconda python=3.10 viralverify hmmer
conda activate viralverify

Install the database:

wget -O nbc_hmms.hmm.gz https://ndownloader.figshare.com/files/17904323
gunzip nbc_hmms.hmm.gz

Usage

viralverify -f input.fa -o outpath --hmm nbc_hmms.hmm -t 8 -p

Notes

-f : Input FASTA file (contigs kept in Filter 1)
-o : Output directory
--hmm : HMM database (decompressed)
-t : Number of threads
-p : Writes predicted plasmid contigs to a separate FASTA file

Filter 3 — Second k-mer classification (PlasFlow)

Classifies contigs as plasmid or chromosome with a neural network based on k-mer frequency signatures. Contigs labelled plasmid are kept.

Setup

conda create -y -n plasflow -c conda-forge -c bioconda plasflow
conda activate plasflow
conda install -y -c conda-forge "libgfortran=3"
pip install --no-deps --force-reinstall protobuf==3.6.1

Usage

PlasFlow.py --input input.fa --output output.tsv --threshold 0.7

Notes

--input : Input FASTA file (contigs kept in Filter 2)
--output : Output table. FASTA files of the classified sequences (_plasmids, _chromosomes, _unclassified) are written next to it.
--threshold : Minimum class probability for a label; contigs below it are unclassified

Filter 4 — rRNA content removal (Infernal)

Detects rRNA genes with Rfam covariance models. A hit is counted when it covers ≥ 30% of the model's consensus length. Contigs with at least one hit are removed.

Setup

conda create -y -n infernal -c conda-forge -c bioconda infernal
conda activate infernal

Install the Rfam rRNA covariance models:

Rfam ID	Rfam family	rRNA
RF00001	5S_rRNA	5S rRNA
RF00177	SSU_rRNA_bacteria	Bacterial 16S rRNA
RF02541	LSU_rRNA_bacteria	Bacterial 23S rRNA
RF01959	SSU_rRNA_archaea	Archaeal 16S rRNA
RF02540	LSU_rRNA_archaea	Archaeal 23S rRNA

RFAM_IDS=(RF00001 RF00177 RF02541 RF01959 RF02540)

for id in "${RFAM_IDS[@]}"; do
    curl -sfL "https://rfam.org/family/${id}/cm"
done > rrna.cm

cmpress -F rrna.cm

Usage

cmscan --noali \
       --cut_ga \
       -g \
       --nohmmonly \
       --rfam \
       --cpu 4 \
       --tblout output.tblout \
       rrna.cm \
       input.fa

Notes

input.fa : Input FASTA file (contigs kept in Filter 3)
--tblout : Tabular output file of the hits
--cpu : Number of threads
--cut_ga : Uses each Rfam family's gathering threshold as the reporting cutoff
-g : Glocal mode (the full model is aligned within the contig)
--nohmmonly : Scores with the covariance models, not only the profile-HMM stage
--rfam : Rfam-tuned preset for fast searches
--noali : Omits alignments from the output

Filter 5 — Human-contamination removal (minimap2)

Aligns contigs to the human reference genome (GRCh38.p14, RefSeq GCF_000001405.40). Contigs with identity ≥ 95% and coverage ≥ 80% are removed. The remaining contigs form the refined plasmid set.

Setup

conda create -y -n minimap2 -c conda-forge -c bioconda minimap2 bedtools samtools
conda activate minimap2

Download the reference genome:

wget https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/001/405/GCF_000001405.40_GRCh38.p14/GCF_000001405.40_GRCh38.p14_genomic.fna.gz
zcat GCF_000001405.40_GRCh38.p14_genomic.fna.gz > GRCh38.p14.fna

Usage

minimap2 -x asm20 -c --secondary=no -I 1G -t 8 GRCh38.p14.fna input.fa > output.paf

Notes

input.fa : Input FASTA file (contigs kept in Filter 4)
-x asm20 : Preset for contig-to-genome alignment (up to ~5% divergence)
-c : Computes base-level alignments
--secondary=no : Reports only primary and supplementary alignments
-I 1G : Loads the index in 1 Gbp parts
-t : Number of threads

References

Q. Zhu, S. Gao, B. Xiao, Z. He, and S. Hu, “Plasmer: an Accurate and Sensitive Bacterial Plasmid Prediction Tool Based on Machine Learning of Shared k-mers and Genomic Features”, Microbiol. Spectr., vol. 11, no. 3, p. e0464522, Jun. 2023, doi: 10.1128/spectrum.04645-22. Github: https://github.com/nekokoe/Plasmer
R. Chikhi et al., “Logan: Planetary-Scale Genome Assembly Surveys Life’s Diversity”, bioRxiv, p. 2024.07.30.605881, Sep. 2025, doi: 10.1101/2024.07.30.605881. Github: https://github.com/IndexThePlanet/Logan
W. Shen, B. Sipos, and L. Zhao, “SeqKit2: A Swiss army knife for sequence and alignment processing”, iMeta, vol. 3, no. 3, p. e191, Jun. 2024, doi: 10.1002/imt2.191. Github: https://github.com/shenwei356/seqkit
M. Steinegger and J. Söding, “MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets”, Nat. Biotechnol., vol. 35, no. 11, pp. 1026-1028, Nov. 2017, doi: 10.1038/nbt.3988. Github: https://github.com/soedinglab/MMseqs2
D. Hyatt, G.-L. Chen, P. F. LoCascio, M. L. Land, F. W. Larimer, and L. J. Hauser, “Prodigal: prokaryotic gene recognition and translation initiation site identification”, BMC Bioinformatics, vol. 11, p. 119, Mar. 2010, doi: 10.1186/1471-2105-11-119. Github: https://github.com/hyattpd/prodigal
M. G. Durrant and A. S. Bhatt, “Automated Prediction and Annotation of Small Open Reading Frames in Microbial Genomes”, Cell Host Microbe, vol. 29, no. 1, pp. 121-131.e4, Jan. 2021, doi: 10.1016/j.chom.2020.11.002. Github: https://github.com/bhattlab/SmORFinder
C. P. Cantalapiedra, A. Hernández-Plaza, I. Letunic, P. Bork, and J. Huerta-Cepas, “eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale”, Mol. Biol. Evol., vol. 38, no. 12, pp. 5825-5829, Dec. 2021, doi: 10.1093/molbev/msab293. Github: https://github.com/eggnogdb/eggnog-mapper
M. Feldgarden et al., “AMRFinderPlus and the Reference Gene Catalog facilitate examination of the genomic links among antimicrobial resistance, stress response, and virulence”, Sci. Rep., vol. 11, no. 1, p. 12728, Jun. 2021, doi: 10.1038/s41598-021-91456-0. Github: https://github.com/ncbi/amr
B. P. Alcock et al., “CARD 2023: expanded curation, support for machine learning, and resistome prediction at the Comprehensive Antibiotic Resistance Database”, Nucleic Acids Res., vol. 51, no. D1, pp. D690-D699, Jan. 2023, doi: 10.1093/nar/gkac920. Github: https://github.com/arpcard/rgi
J. Zheng, Q. Ge, Y. Yan, X. Zhang, L. Huang, and Y. Yin, “dbCAN3: automated carbohydrate-active enzyme and substrate annotation”, Nucleic Acids Res., vol. 51, no. W1, pp. W115-W121, Jul. 2023, doi: 10.1093/nar/gkad328. Github: https://github.com/bcb-unl/run_dbcan
J. Russel, R. Pinilla-Redondo, D. Mayo-Muñoz, S. A. Shah, and S. J. Sørensen, “CRISPRCasTyper: Automated Identification, Annotation, and Classification of CRISPR-Cas Loci”, CRISPR J., vol. 3, no. 6, pp. 462-469, Dec. 2020, doi: 10.1089/crispr.2020.0059. Github: https://github.com/Russel88/CRISPRCasTyper
Santos-Júnior CD, Pan S, Zhao XM, Coelho LP. "Macrel: antimicrobial peptide screening in genomes and metagenomes." PeerJ. 2020 Dec 18;8:e10555. doi: 10.7717/peerj.10555. Github: https://github.com/BigDataBiology/macrel
Erözden AA, Tavşanlı N, Demirel G, Sanli NO, Çalışkan M, Arıkan M. "MetaPepticon: automated prediction of anticancer peptides from microbial genomes and metagenomes." PeerJ. 2026 Mar 27;14:e20990. doi: 10.7717/peerj.20990. Github: https://github.com/arikanlab/MetaPepticon
D. Pellow, I. Mizrahi, and R. Shamir, “PlasClass improves plasmid sequence classification”, PLoS Comput. Biol., vol. 16, no. 4, p. e1007781, Apr. 2020, doi: 10.1371/journal.pcbi.1007781. Github: https://github.com/Shamir-Lab/PlasClass
D. Antipov, M. Raiko, A. Lapidus, and P. A. Pevzner, “Metaviral SPAdes: assembly of viruses from metagenomic data”, Bioinformatics, vol. 36, no. 14, pp. 4126-4129, Jul. 2020, doi: 10.1093/bioinformatics/btaa490. Github: https://github.com/ablab/viralVerify
P. S. Krawczyk, L. Lipinski, and A. Dziembowski, “PlasFlow: predicting plasmid sequences in metagenomic data using genome signatures”, Nucleic Acids Res., vol. 46, no. 6, p. e35, Apr. 2018, doi: 10.1093/nar/gkx1321. Github: https://github.com/smaegol/PlasFlow
E. P. Nawrocki and S. R. Eddy, “Infernal 1.1: 100-fold faster RNA homology searches”, Bioinformatics, vol. 29, no. 22, pp. 2933-2935, Nov. 2013, doi: 10.1093/bioinformatics/btt509. Github: https://github.com/EddyRivasLab/infernal
N. Ontiveros-Palacios et al., “Rfam 15: RNA families database in 2025”, Nucleic Acids Res., vol. 53, no. D1, pp. D258-D267, Jan. 2025, doi: 10.1093/nar/gkae1023. Website: https://rfam.org
H. Li, “Minimap2: pairwise alignment for nucleotide sequences”, Bioinformatics, vol. 34, no. 18, pp. 3094-3100, Sep. 2018, doi: 10.1093/bioinformatics/bty191. Github: https://github.com/lh3/minimap2

Name		Name	Last commit message	Last commit date
README.md		README.md

Latest commit

History

Folders and files

Repository files navigation

PLATE: Plasmid Atlas of Food Ecosystems

Table of contents

Overview

Step-by-Step Workflow

Part 1 — Data acquisition

LOGAN

Usage

Notes

Assembly (optional)

Cuttlefish2

Setup

Usage

Notes

Minia3

Setup

Usage

Notes

Part 2 — Plasmid prediction

Plasmer

Setup

Install the databases

Usage

Notes

Part 3 — Dereplication and clustering

seqkit

Setup

Usage

Notes

MMseqs2

Setup

Usage

Notes

Part 4 — Gene prediction

Prodigal

Setup

Usage

Notes

smORFinder

Setup

Usage

Notes

Part 5 — Host assignment

HOTSPOT

Setup

Usage

Notes

Part 6 — Functional annotation

eggNOG-mapper

Setup

Usage

Notes

AMRFinderPlus

Setup

Usage

Notes

Resistance Gene Identifier

Setup

Usage

Notes

dbCAN

Setup

Usage

Notes

CRISPR-Cas Typer

Setup

Usage

Notes

Macrel

Setup

Usage

Notes

MetaPepticon

Setup

Usage

Notes

Part 7 — Constructing the refined plasmid set