De novo identification of miRNAs. It is an experimental approach designed to identify miRNAs in small RNA-seq data from species with no genome available to be used as reference.
It was not thoroughly tested and we give no warranty of confident results, please use it with caution.
- collapse 100% identical reads
- filter out tRNA and rRNA reads
- miRNA identification
- cluster similar reads (isoforms with 98% identity)
- detect the star reads (with 90% identity)
- identify known and putative novel miRNAs
If the user has all requirements installed an properly working, you just need to do the following steps (please, change /.bash_profile to /.bashrc if it is your bash source):
git clone https://github.com/pedronachtigall/denomiR.git
echo "export PATH=$PATH:$(pwd)/denomiR/bin/" >> ~/.bash_profile
source ~/.bash_profile
Alternatively, the user can install all requirements through conda manager as follow (please, change /.bash_profile to /.bashrc if it is your bash source):
conda create --name denomiR_env -c bioconda python=3.7 biopython numpy matplotlib cd-hit blast
git clone https://github.com/pedronachtigall/denomiR.git
echo "export PATH=$PATH:$(pwd)/denomiR/bin/" >> ~/.bash_profile
source ~/.bash_profile
conda activate denomiR_env
It may be needed to apply "execution permission" to all bin executables: chmod +x denomiR/bin/*
Usage: denomiR.py [options]
Options:
-h, --help show this help message and exit
-i string, --input=string
Mandatory - reads clipped in fq format (it can be
compressed in gz format also)
-o path, --output=path
Optional - path to output folder [default =
"dnm_output"]
-d path, --db=path Mandatory - path/to/folder/with/DBs, folder containing
the RFAMdb and miRNAs from miRBase and MirGeneDB
-c boolean, --CollapseReads=boolean
Optional - Turn On/Off the collpase reads step with
True/False. If your reads are in fasta format and were
collapsed elsewhere [default = True]
-r boolean, --rfam=boolean
Optional - Turn On/Off the rfam filtering step with
True/False. [default = True]
-l int, --MinLength=int
Optional - Minimum length of reads to used in the
analysis [default = 19]
-L int, --MaxLength=int
Optional - Maximum length of reads to used in the
analysis [default = 25]
-R int, --ReadsDepth=int
Optional - Minimum number of reads to be kept in the
analysis [default = 10]
-s float, --similarity=float
Optional - Similarity presented by reads to be
clustered together. The number must be between 0.1 and
1.0 [default = 0.98]
Basic usage:
denomiR.py -i reads_clipped.fq(.gz) -o output_folder -db path/to/rfam&mirnaDB
To plot charts for each miRNA, please use the plot_function.py script as follow:
plot_function.py known_mirnas.fa novel_mirnas.fa mirna_clusters.txt output_folder
The denomiR was designed to work with the clipped reads in fastq format as input. The file can be compressed in gzip format. Please notice, that the adapter must be trimmed. The user must use any tool available to perform this task.
The user must also indicate the PATH to the RFAM and miRNAdb -db path/to/denomiR/mirDB/.
Please, notice that the current DB is available in a zip file format. Decompress this file before running denomiR with the command unzip mirDB.zip.
The current DB available is designed for Metazoa species, but the user can design a DB specific to their lineage and replace the files in the mirDB folder.
By default, all files generated during the analysis are kept. The final_results.txt is the final summary of results and containing read counts and the mature and star sequences identified for known and putative novel miRNAs. The sequences of miRNAs identified can be found in the fasta files known_mirnas.fa and novel_mirnas.fa. If the user also uses the plot_function.py a folder named "plots" will contain charts with the expression data in each position of the miRNA for each known and putative novel miRNA identified in the analysis.
denomiR_output/
├── cdhit.clstr
├── final_results.txt
├── known_mirnas.fa
├── mirna_clusters.txt
├── mirna_seq.fa
├── novel_mirnas.fa
├── other_seqs.fa
├── rfam_blast.out
└── plots/
├── known/
| ├── id_1.pdf
| ├── ...
| └── id_N.pdf
└── novel/
├── id_1.pdf
├── ...
└── id_N.pdf
🐛🆘💬
To report bugs, to ask for help and to give any feedback, please contact Pedro G. Nachtigall: pedronachtigall@gmail.com
If you use denomiR in your analysis, please cite the following (which is the manuscript that denomiR was designed to be used):
Nachtigall et al. (2022) Differences in PLA2 Constitution Distinguish the Venom of Two Endemic Brazilian Mountain Lanceheads, Bothrops cotiara and Bothrops fonsecai. Toxins, 14(4), 237. DOI:https://doi.org/10.3390/toxins14040237