[](https://doi.org/10.5281/zenodo.5268144)
# DRFP
An NLP-inspired chemical reaction fingerprint based on basic set arithmetic.
Read the associated [open access article](https://pubs.rsc.org/en/content/articlehtml/2022/dd/d1dd00006c)
## Description
Predicting the nature and outcome of reactions using computational methods is an important tool to accelerate chemical research. The recent application of deep learning-based learned fingerprints to reaction classification and reaction yield prediction has shown an impressive increase in performance compared to previous methods such as DFT- and structure-based fingerprints. However, learned fingerprints require large training data sets, are inherently biased, and are based on complex deep learning architectures. Here we present the differential reaction fingerprint *DRFP*. The *DRFP* algorithm takes a reaction SMILES as an input and creates a binary fingerprint based on the symmetric difference of two sets containing the circular molecular n-grams generated from the molecules listed left and right from the reaction arrow, respectively, without the need for distinguishing between reactants and reagents. We show that *DRFP* outperforms DFT-based fingerprints in reaction yield prediction and other structure-based fingerprints in reaction classification, and reaching the performance of state-of-the-art learned fingerprints in both tasks while being data-independent.
## Getting Started
The best way to start exploring DRFP is on binder.
A notebook that gets you started on creating and using DRFP:
[](https://mybinder.org/v2/gh/reymond-group/drfp/HEAD?filepath=notebooks%2F01_fingerprinting.ipynb)
A notbook that explains how you can use SHAP to analyse and interpret your machine learning models when using DRFP:
[](https://mybinder.org/v2/gh/reymond-group/drfp/HEAD?filepath=notebooks%2F02_model_explainability.ipynb)
## Installation and Usage
*DRFP* can be installed from pypi using `pip install drfp`. However, it depends on [RDKit](https://www.rdkit.org/) which is best [installed using conda](https://www.rdkit.org/docs/Install.html).
Once DRFP is installed, there are two ways you can use it. You can use the cli app `drfp` or the library provided by the package.
### CLI
```bash
drfp my_rxn_smiles.txt my_rxn_fps.pkl -d 512
```
This will create a pickle dump containing an numpy ndarray containing DRFP fingerprints with a dimensionality of 512. To also export the mapping, use the flag `--mapping`. This will create the additional file `my_rxn_fps.map.pkl`. You can call `drfp --help` to show all available flags and options.
### Library
Following is a basic exmple of how to use DRFP in a Python script.
```python
from drfp import DrfpEncoder
rxn_smiles = [
"CO.O[C@@H]1CCNC1.[C-]#[N+]CC(=O)OC>>[C-]#[N+]CC(=O)N1CC[C@@H](O)C1",
"CCOC(=O)C(CC)c1cccnc1.Cl.O>>CCC(C(=O)O)c1cccnc1",
]
fps = DrfpEncoder.encode(rxn_smiles)
```
The variable `fps` now points to a list containing the fingerprints for the two reaction SMILES as numpy arrays.
## Documentation
The library contains the class `DrfpEncoder` with one public method `encode`.
| `DrfpEncoder.encode()` | Description | Type | Default |
|-|-|-|-|
| `X` | An iterable (e.g. a list) of reaction SMILES or a single reaction SMILES to be encoded | `Iterable` or `str` | |
| `n_folded_length` | The folded length of the fingerprint (the parameter for the modulo hashing) | `int` | `2048` |
| `min_radius` | The minimum radius of a substructure (0 includes single atoms) | `int` | `0` |
| `radius` | The maximum radius of a substructure | `int` | `3` |
| `rings` | Whether to include full rings as substructures | `bool` | `True` |
| `mapping` | Return a feature to substructure mapping in addition to the fingerprints. If true, the return signature of this method is `Tuple[List[np.ndarray], Dict[int, Set[str]]]` | `bool` | `False` |
| `atom_index_mapping` | Return the atom indices of mapped substructures for each reaction | `bool` | `False` |
| `root_central_atom` | Whether to root the central atom of substructures when generating SMILES | `bool` | `True` |
| `include_hydrogens` | Whether to explicitly include hydrogens in the molecular graph | `bool` | `False` |
| `show_progress_bar` | Whether to show a progress bar when encoding reactions | `bool` | `False` |
# Reproduce
Want to reproduce the results in our paper? You can find all the data in the `data` folder and encoding and training scripts in the `scripts` folder.
# Cite Us
```
@article{probst2022reaction,
title={Reaction Classification and Yield Prediction using the Differential Reaction Fingerprint DRFP},
author={Probst, Daniel and Schwaller, Philippe and Reymond, Jean-Louis},
journal={Digital Discovery},
year={2022},
publisher={Royal Society of Chemistry}
}
```
Raw data
{
"_id": null,
"home_page": "https://github.com/reymond-group/drfp",
"name": "drfp",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "",
"author": "Daniel Probst",
"author_email": "daniel.probst@hey.com",
"download_url": "https://files.pythonhosted.org/packages/84/8d/f266aabfadbf5547e8da47457dd01c5e147ffd3978fd6eb0de348b02bc81/drfp-0.3.6.tar.gz",
"platform": "any",
"description": "\n[](https://doi.org/10.5281/zenodo.5268144)\n\n\n# DRFP\n\n\nAn NLP-inspired chemical reaction fingerprint based on basic set arithmetic.\n\nRead the associated [open access article](https://pubs.rsc.org/en/content/articlehtml/2022/dd/d1dd00006c)\n\n\n## Description\n\nPredicting the nature and outcome of reactions using computational methods is an important tool to accelerate chemical research. The recent application of deep learning-based learned fingerprints to reaction classification and reaction yield prediction has shown an impressive increase in performance compared to previous methods such as DFT- and structure-based fingerprints. However, learned fingerprints require large training data sets, are inherently biased, and are based on complex deep learning architectures. Here we present the differential reaction fingerprint *DRFP*. The *DRFP* algorithm takes a reaction SMILES as an input and creates a binary fingerprint based on the symmetric difference of two sets containing the circular molecular n-grams generated from the molecules listed left and right from the reaction arrow, respectively, without the need for distinguishing between reactants and reagents. We show that *DRFP* outperforms DFT-based fingerprints in reaction yield prediction and other structure-based fingerprints in reaction classification, and reaching the performance of state-of-the-art learned fingerprints in both tasks while being data-independent.\n\n## Getting Started\nThe best way to start exploring DRFP is on binder.\nA notebook that gets you started on creating and using DRFP:\n\n[](https://mybinder.org/v2/gh/reymond-group/drfp/HEAD?filepath=notebooks%2F01_fingerprinting.ipynb)\n\nA notbook that explains how you can use SHAP to analyse and interpret your machine learning models when using DRFP:\n\n[](https://mybinder.org/v2/gh/reymond-group/drfp/HEAD?filepath=notebooks%2F02_model_explainability.ipynb)\n\n## Installation and Usage\n*DRFP* can be installed from pypi using `pip install drfp`. However, it depends on [RDKit](https://www.rdkit.org/) which is best [installed using conda](https://www.rdkit.org/docs/Install.html).\n\nOnce DRFP is installed, there are two ways you can use it. You can use the cli app `drfp` or the library provided by the package.\n\n### CLI\n```bash\ndrfp my_rxn_smiles.txt my_rxn_fps.pkl -d 512\n```\n\nThis will create a pickle dump containing an numpy ndarray containing DRFP fingerprints with a dimensionality of 512. To also export the mapping, use the flag `--mapping`. This will create the additional file `my_rxn_fps.map.pkl`. You can call `drfp --help` to show all available flags and options.\n\n### Library\nFollowing is a basic exmple of how to use DRFP in a Python script.\n```python\nfrom drfp import DrfpEncoder\n\nrxn_smiles = [\n \"CO.O[C@@H]1CCNC1.[C-]#[N+]CC(=O)OC>>[C-]#[N+]CC(=O)N1CC[C@@H](O)C1\",\n \"CCOC(=O)C(CC)c1cccnc1.Cl.O>>CCC(C(=O)O)c1cccnc1\",\n]\n\nfps = DrfpEncoder.encode(rxn_smiles)\n```\n\nThe variable `fps` now points to a list containing the fingerprints for the two reaction SMILES as numpy arrays.\n\n\n## Documentation\n\nThe library contains the class `DrfpEncoder` with one public method `encode`.\n\n| `DrfpEncoder.encode()` | Description | Type | Default |\n|-|-|-|-|\n| `X` | An iterable (e.g. a list) of reaction SMILES or a single reaction SMILES to be encoded | `Iterable` or `str` | |\n| `n_folded_length` | The folded length of the fingerprint (the parameter for the modulo hashing) | `int` | `2048` |\n| `min_radius` | The minimum radius of a substructure (0 includes single atoms) | `int` | `0` |\n| `radius` | The maximum radius of a substructure | `int` | `3` |\n| `rings` | Whether to include full rings as substructures | `bool` | `True` |\n| `mapping` | Return a feature to substructure mapping in addition to the fingerprints. If true, the return signature of this method is `Tuple[List[np.ndarray], Dict[int, Set[str]]]` | `bool` | `False` |\n| `atom_index_mapping` | Return the atom indices of mapped substructures for each reaction | `bool` | `False` |\n| `root_central_atom` | Whether to root the central atom of substructures when generating SMILES | `bool` | `True` |\n| `include_hydrogens` | Whether to explicitly include hydrogens in the molecular graph | `bool` | `False` |\n| `show_progress_bar` | Whether to show a progress bar when encoding reactions | `bool` | `False` |\n\n# Reproduce\nWant to reproduce the results in our paper? You can find all the data in the `data` folder and encoding and training scripts in the `scripts` folder.\n\n# Cite Us\n```\n@article{probst2022reaction,\n title={Reaction Classification and Yield Prediction using the Differential Reaction Fingerprint DRFP},\n author={Probst, Daniel and Schwaller, Philippe and Reymond, Jean-Louis},\n journal={Digital Discovery},\n year={2022},\n publisher={Royal Society of Chemistry}\n}\n```\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "An NLP-inspired chemical reaction fingerprint based on basic set arithmetic.",
"version": "0.3.6",
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"md5": "bab7e695604ba3c59acadf1259574579",
"sha256": "061cd7a3ea2cbdaefff5b89b56da2925cc56ca4220f43ef03565e86dbacd69af"
},
"downloads": -1,
"filename": "drfp-0.3.6-py2.py3-none-any.whl",
"has_sig": false,
"md5_digest": "bab7e695604ba3c59acadf1259574579",
"packagetype": "bdist_wheel",
"python_version": "py2.py3",
"requires_python": null,
"size": 9450,
"upload_time": "2022-12-30T20:18:52",
"upload_time_iso_8601": "2022-12-30T20:18:52.369813Z",
"url": "https://files.pythonhosted.org/packages/0c/1c/959d25a4db04040463c45c081a14f25722b4e19bd26df62a1d21dea18739/drfp-0.3.6-py2.py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"md5": "c46cb4d02ebc744923a237b9c3692eca",
"sha256": "987a8bc36537817d02940618d078817e2891ad499abd1965bf0aacdcb73c5d83"
},
"downloads": -1,
"filename": "drfp-0.3.6.tar.gz",
"has_sig": false,
"md5_digest": "c46cb4d02ebc744923a237b9c3692eca",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 94628170,
"upload_time": "2022-12-30T20:18:59",
"upload_time_iso_8601": "2022-12-30T20:18:59.897577Z",
"url": "https://files.pythonhosted.org/packages/84/8d/f266aabfadbf5547e8da47457dd01c5e147ffd3978fd6eb0de348b02bc81/drfp-0.3.6.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2022-12-30 20:18:59",
"github": true,
"gitlab": false,
"bitbucket": false,
"github_user": "reymond-group",
"github_project": "drfp",
"travis_ci": false,
"coveralls": true,
"github_actions": true,
"tox": true,
"lcname": "drfp"
}
JSON
