Publication: Torres, Maria Celeste, et al. "In Silico Analysis of Dengue Virus Serotype 2 Mutations Detected at the Intrahost Level in Patients with Different Clinical Outcomes." Microbiology Spectrum 9.2 (2021): e00256-21. DOI: https://doi.org/10.1128/Spectrum.00256-21.
Install dependencies using conda:
conda env create -f environment.yml
First, cd modellr.
To generate models, you will need at least two inputs:
- target sequence in FASTA format;
- template structure in PDB format.
Target sequences must be converted to PIR format before proceeding. A script fasta2pir.py for simple cases is available.
Usage:
fasta2pir.py [-h] [-f FASTA] [-o OUTPUT]
Very basic script to convert FASTA to PIR format.
arguments:
-h, --help show this help message and exit
-f FASTA, --fasta FASTA
fasta file with one sequence
-o OUTPUT, --output OUTPUT
output path
Example:
python fasta2pir.py --fasta files/basic-example/ns3#123.fasta --output files/basic-example/
All set, you can build models using the run.sh script, which uses align2d.py, build_model.py, evaluate_model.py, and also runs PROCHECK with the best model obtained as input, if the software is installed.
To use the bash script:
bash.sh [--template TEMPLATE] [--target TARGET] [--chains CHAINS] [--dir DIR]
arguments:
--template TEMPLATE
PDB file of the template
--target TARGET
Target sequence file (in PIR format, .ali)
--chains CHAINS
First and last chain to be used (e.g. AB). In case of using just one chain, repeat it's letter (e.g. AA)
--dir DIR
Path to dir containing input files
Example:
bash run.sh --target ns3#123.ali --template 5yw1.pdb --chains BB --dir files/basic-example
This script generates several output files and logs (see the MODELLER manual for details), including alignment files (.ali and .pap) and a quality assessment plot such as the one below. Also, a folder named procheck is created containing the PROCHECK results.
