Carbon-based Multi-level AtomicNeiGhborhOod EncodingS.
This package is created for the following paper:
"A parametric approach for molecular encodings using multilevel atomic neighborhoods applied to peptide classification" by Georges Hattab, Aleksandar Anžel, Sebastian Spänig, Nils Neumann, Dominik Heider
Please cite the paper as:
@article{10.1093/nargab/lqac103,
author = {Hattab, Georges and Anžel, Aleksandar and Spänig, Sebastian and Neumann, Nils and Heider, Dominik},
title = "{A parametric approach for molecular encodings using multilevel atomic neighborhoods applied to peptide classification}",
journal = {NAR Genomics and Bioinformatics},
volume = {5},
number = {1},
year = {2023},
month = {01},
abstract = "{Exploring new ways to represent and discover organic molecules is critical to the development of new therapies. Fingerprinting algorithms are used to encode or machine-read organic molecules. Molecular encodings facilitate the computation of distance and similarity measurements to support tasks such as similarity search or virtual screening. Motivated by the ubiquity of carbon and the emerging structured patterns, we propose a parametric approach for molecular encodings using carbon-based multilevel atomic neighborhoods. It implements a walk along the carbon chain of a molecule to compute different representations of the neighborhoods in the form of a binary or numerical array that can later be exported into an image. Applied to the task of binary peptide classification, the evaluation was performed by using forty-nine encodings of twenty-nine data sets from various biomedical fields, resulting in well over 1421 machine learning models. By design, the parametric approach is domain- and task-agnostic and scopes all organic molecules including unnatural and exotic amino acids as well as cyclic peptides. Applied to peptide classification, our results point to a number of promising applications and extensions. The parametric approach was developed as a Python package (cmangoes), the source code and documentation of which can be found at https://github.com/ghattab/cmangoes and https://doi.org/10.5281/zenodo.7483771.}",
issn = {2631-9268},
doi = {10.1093/nargab/lqac103},
url = {https://doi.org/10.1093/nargab/lqac103},
note = {lqac103},
eprint = {https://academic.oup.com/nargab/article-pdf/5/1/lqac103/48591755/lqac103.pdf},
}
Abstract:
Exploring new ways to represent and discover organic molecules is critical to the development of new therapies. Fingerprinting algorithms are used to encode or machine-read organic molecules. Molecular encodings facilitate the computation of distance and similarity measurements to support tasks such as similarity search or virtual screening. Motivated by the ubiquity of carbon and the emerging structured patterns, we propose a parametric approach for molecular encodings using carbon-based multilevel atomic neighborhoods. It implements a walk along the carbon chain of a molecule to compute different representations of the neighborhoods in the form of a binary or numerical array that can later be exported into an image. Applied to the task of binary peptide classification, the evaluation was performed by using forty-nine encodings of twenty-nine data sets from various biomedical fields, resulting in well over 1421 machine learning models. By design, the parametric approach is domain- and task-agnostic and scopes all organic molecules including unnatural and exotic amino acids as well as cyclic peptides. Applied to peptide classification, our results point to a number of promising applications and extensions. The parametric approach was developed as a Python package (cmangoes), the source code and documentation of which can be found at https://github.com/ghattab/cmangoes and https://doi.org/10.5281/zenodo.7483771.
The code is written in Python 3.8.12 and tested on Linux with the following libraries installed:
| Library | Version |
|---|---|
| biopython | 1.78 |
| ipython | 7.27.0 |
| matplotlib | 3.4.3 |
| networkx | 2.6.3 |
| numpy | 1.21.2 |
| openbabel | 3.1.1 |
| pandas | 1.3.4 |
| pysmiles | 1.0.1 |
The example data we used to evaluate CMANGOES comes from the peptidereactor repository. We selected 30 data sets from that repository and placed them at Data/Original_datasets/. Each data set has a separate README file that contains the additional information of that data set.
After encoding those data sets with CMANGOES, we used the peptidereactor to evaluate our method against all other encoding methods available in the peptidereactor. Evaluation results of CMANGOES are placed at Data/Visualization_data/data/. Evaluation results of other encoding methods provided by the peptidereactor are located at Data/Visualization_data/peptidereactor_vis_data/. We combined both evaluation results to create visualizations present in our manuscript.
| Script | Description |
|---|---|
| Code/ | contains all scripts necessary to run the tool. |
| Code/cmangoes.py | contains the code that implements the whole pipeline. |
| Code/batch_encoding.py | contains the code that uses CMANGOES to batch-encode selected data sets from the peptidereactor repository. |
| Code/visualize.ipynb | contains the code that creates visualizations of the final results. |
| Code/Dimensionality_Reduction.Rmd | contains the code used for ML tasks in the paper, specifically the dimensionality reduction step of ML pipeline. |
| Code/Machine_Learning.Rmd | contains the code used for ML tasks in the paper. |
| Code/Preprocessing.Rmd | contains the code used for ML tasks in the paper, specifically the preprocessing step of ML pipeline. |
usage: cmangoes [-h] [--level {1,2,12}] [--image {0,1}] [--show_graph SHOW_GRAPH]
[--output_path OUTPUT_PATH]
input_file {b,d} {c,s}
cmangoes: Carbon-based Multi-level Atomic Neighborhood Encodings
positional arguments:
input_file A required path-like argument
{b,d} A required character argument that specifies an encoding to be used. b is for
binary, d is for discretized
{c,s} A required character argument that specifies a padding to be used. c is for
centered, s is for shifted
optional arguments:
-h, --help show this help message and exit
--level {1,2,12} An optional integer argument that specifies the upper boundary of levels that
should be considered. Default: 12 (levels 1 and 2). Example: level 1 returns
only first-level neighbors, and level 2 return only second-level neighbors.
Level 12 returns level 1 and level 2 neighbors
--image {0,1} An optional integer argument that specifies whether images should be created or
not. Default: 0 (without images)
--show_graph SHOW_GRAPH
An optional integer argument that specifies whether a graph representation
should be created or not. Default: 0 (without representation). The user should
provide the number between 1 and the number of sequences in the parsed input
file. Example: if number 5 is parsed for this option, a graph representation of
the 5th sequence of the input file shall be created and placed in the
corresponding images folder
--output_path OUTPUT_PATH
An optional path-like argument. For parsed paths, the directory must exist
beforehand. Default: ./CMANGOES_Results
Place yourself in the Code directory, then run the following command python cmangoes.py --help to see how to run the tool. The output of this option is presented in the previous subsection.
You can also use the standalone CMANGOES executable on GNU Linux systems. To do that, place yourself in the Executable directory and then run the tool with the following command ./cmangoes input_file {b, d} {c, s} or just use ./cmangoes --help to see how to run the tool.
Licensed under the MIT License (LICENSE).
Any contribution intentionally submitted for inclusion in the work by you, shall be licensed under the MIT Licence.
