This pipeline simulates metabarcoding sequencing runs (paired-end) according to a user provided sample composition and reference sequence database.
MetaSeqSim runs in a UNIX environment with BASH (tested on Debian GNU/Linux 10 (buster)) and requires conda and an internet connection (at least for the first run).
Start by getting a copy of this repository on your system, either by downloading and unpacking the archive, or using 'git clone':
cd path/to/repo/
git clone https://github.com/CVUA-RRW/MetaSeqSimSet up a conda environment containing snakemake, python and the pandas library and activate it:
conda create --name snakemake -c bioconda -c anaconda snakemake pandas
conda activate snakemakeTo run the pipeline you will need to provide a BLAST-formated reference sequence database. If you already have a fasta file with your sequences follow the BLAST documentation to know how to format it.
If you want to extract barcodes from a database of reference genomes you can check out our RRW-PrimerBLAST pipeline.
You will also need to provide the taxdb files available from the NCBI server.
MetaSeqSim should be run using the snakemake command-line application. For this you will need to manually fill the config.yaml file with the paths to the required files. You can also modify the parameters already present in the file.
Then run the pipeline with:
snakemake -s /path/to/MetaSeqSim/Snakefile --configfile path/to/config.yaml --use-conda --conda-prefix path/to/your/conda/envsConsult snakemake's documentation for more details.
You can enter the sample composition in a tabular file, as shown below. Enter one line per species, repeating the sample name to combine species in the same sample. Species are to be provided with their taxonomic node number according to the NCBI's taxonomic classififcation.
Note the the read count values affects both reads: a value of 10000 will generate 10000 R1 reads and 10000 R2 reads.
| sample | taxid | read_count |
|---|---|---|
| sample1 | 9913 | 250000 |
| sample1 | 9903 | 100000 |
| sample1 | 9906 | 200 |
| sample2 | 9906 | 500000 |
| sample2 | 9923 | 500 |
Be sure to use tabulations as field separator and to respect column naming!
The configuration file contains the following parameters:
# Fill in the path belows with your own specifications:
workdir: # Path to output directory
blast_db: # Path to BLAST-formated database
taxdb: # Path to the folder containing the taxdb files
samples: # Path to the sample definition table
# Modify the parameters below:
read_length: 150 # Sequencing read length
mean_fragment_size: 151 # Mean DNA fragment size. Must be > than read_length
quality_shift_R2: 0 # Quality shift of read 2
fragment_size_sdev: 0 # Standard deviation of fragment size
seq_system: MSv3 # Sequencing system, ignored if providing custom profiles
quality_profile_read1: None # Custom quality profile for read1
quality_profile_read2: None # Custom quality profile for read2
If the reference sequence is shorter than the provided read_length and mean_fragment_length, these will be reduced to math the sequence size, resulting in a full fragment sequencing read.
Accepted values for seq_system are :
GA1 - GenomeAnalyzer I (36bp,44bp), GA2 - GenomeAnalyzer II (50bp, 75bp)
HS10 - HiSeq 1000 (100bp), HS20 - HiSeq 2000 (100bp), HS25 - HiSeq 2500 (125bp, 150bp)
HSXn - HiSeqX PCR free (150bp), HSXt - HiSeqX TruSeq (150bp), MinS - MiniSeq TruSeq (50bp)
MSv1 - MiSeq v1 (250bp), MSv3 - MiSeq v3 (250bp), NS50 - NextSeq500 v2 (75bp)
It is better to use custom quality profiles rather than the buit-in ones.
You can create such profiles based on previous sequencing runs by directly using the built-in
ART command art_profiler_illumina.
Alternatively you can use the convenience script scripts/create_error_profile.sh
of this repo. The script will take care of renaming your sequencing files to match
the requirements of ART.
Use it as follows:
conda create --name ART art
conda activate ART
bash /path/to/MetaSeqSim/scripts/create_error_profile.sh -o /path/to/out/dir -f /path/to/sequencing_dir -o PROFILE_NAME_
This will create two files named PROFILE_NAME_R1.txt and PROFILE_NAME_R2.txt that you can for the quality_profile parameters in the config file.
The Pipeline will produce two files per samples, following the illumina naming convention:
- <sample_name>_S1_L001_R1_001.fastq.gz
- <sample_name>_S1_L001_R2_001.fastq.gz
The reference sequences that were used to genrate these files will also be outputed in individual fasta files for inspection.
MetaSeqSim is built with Snakemake and uses:
For new features or to report bugs please submit issues directly on the online repository.
This project is licensed under a BSD 3-Clauses License, see the LICENSE file for details.
For questions about the pipeline, problems, suggestions or requests, feel free to contact:
Grégoire Denay, Chemisches- und Veterinär-Untersuchungsamt Rhein-Ruhr-Wupper