frustratometeR

An R package for calculating energetic local frustration in protein structures. Additionally to the functionalities present at the web server, frustratometeR offers the possibility to analyse frustration across molecular dynamics simulations and to assess the effect of aminoacid variants.

Installation

invisible(lapply(c("usethis", "devtools"), library, character.only = TRUE))

devtools::install_github("proteinphysiologylab/frustratometeR")

ONLY for UBUNTU 22.04 and for R version 4.X.X: To install frustratometeR from the bash file download the install_frustraR.sh file and in a terminal run the follow command line: sudo sh install_frustraR.sh (the installation of modeller is not included in this file)

Dependencies

** If you are using MacOs, FrustratometeR is compatible with systems using MacOS version 10.15 or higher.

** If you are using UBUNTU, FrustratometeR is compatible with Ubuntu 16 or higher.

R packages (R version >= 3.6.3 required. Compatible with R 4.0.X)

install.packages(c("usethis", "devtools"))

Python 3

You will need Python 3.7 (3.9 preferred)

• numpy python3 -m pip install numpy

• biopython python3 -m pip install biopython

• leiden(optional) python3 -m pip install leidenalg

others

• pymol sudo apt install pymol

• magick sudo apt-get install -y libmagick++-dev

• perl sudo apt-get install perl

modeller (you need to install a licenced version of modeller in order to use some of the frustratometeR functionalities, https://salilab.org/modeller/, we recommend to install the latest release) modeller needs to be available via python3. To check if this is working execute python3 and try to do "from modeller import * " after installing modeller. We suggest to install modeller using conda environments.

if modeller does not work from RStudio, you can use reticulate to tell R which conda environment to use. Execute these lines before you execute the frustration functions.

library(reticulate)

use_python("/url/bin/python3")

use_condaenv("condaenvironment")

Sys.setenv(RETICULATE_PYTHON = "/url/bin/python3")

reticulate::py_config()

you have to replace "/url/bin/python3" with the location of your python. You can get that location by doing which python3 in a system terminal.

you have to replace "condaenvironment" with the conda environment where you installed modeller.

It also might be needed to export the environment variables:

find /usr/lib/modellerXX.X/ -name ”modeller”

find /usr/lib/modellerXX.X/ -name ”_modeller.so”

export PYTHONPATH:$PYTHONPATH:/path/to/modeller/

export PYTHONPATH:$PYTHONPATH:/path/to/_modeller/

Minimum code to calculate frustration in a protein

library(frustratometeR)

Pdb_conf <- calculate_frustration(PdbID = "1n0r",Chain = "A", ResultsDir = "/Home/Desktop/")

view_frustration_pymol(Pdb_conf)

You can find an example of how to use the package at:

https://github.com/proteinphysiologylab/frustratometeR/tree/master/Examples

You can also find useful examples in our wiki!!:

https://github.com/proteinphysiologylab/frustratometeR/wiki

You can find the data used in the Wiki examples at:

https://github.com/proteinphysiologylab/frustratometer-data.git

Google Colab Notebook to try FrustratometeR online

https://github.com/proteinphysiologylab/frustratometeR/blob/master/FrustratometeR_Colab.ipynb