SMILES2IUPAC-canonical-small

SMILES2IUPAC-canonical-small

SMILES2IUPAC-canonical-small was designed to accurately translate SMILES chemical names to IUPAC standards.

Model Details

Model Description

SMILES2IUPAC-canonical-small is based on the MT5 model with optimizations in implementing different tokenizers for the encoder and decoder.

• Developed by: Knowladgator Engineering
• Model type: Encoder-Decoder with attention mechanism
• Language(s) (NLP): SMILES, IUPAC (English)
• License: Apache License 2.0

Model Sources

• Paper: coming soon
• Demo: ChemicalConverters

Quickstart

Firstly, install the library:

pip install chemical-converters

SMILES to IUPAC

! Preferred IUPAC style

To choose the preferred IUPAC style, place style tokens before your SMILES sequence.

To perform simple translation, follow the example:

from chemicalconverters import NamesConverter

converter = NamesConverter(model_name="knowledgator/SMILES2IUPAC-canonical-small")
print(converter.smiles_to_iupac('CCO'))
print(converter.smiles_to_iupac(['<SYST>CCO', '<TRAD>CCO', '<BASE>CCO']))

['ethanol']
['ethanol', 'ethanol', 'ethanol']

Processing in batches:

from chemicalconverters import NamesConverter

converter = NamesConverter(model_name="knowledgator/SMILES2IUPAC-canonical-small")
print(converter.smiles_to_iupac(["<BASE>C=CC=C" for _ in range(10)], num_beams=1, 
                                process_in_batch=True, batch_size=1000))

['buta-1,3-diene', 'buta-1,3-diene'...]

Validation SMILES to IUPAC translations

It's possible to validate the translations by reverse translation into IUPAC and calculating Tanimoto similarity of two molecules fingerprints.

from chemicalconverters import NamesConverter

converter = NamesConverter(model_name="knowledgator/SMILES2IUPAC-canonical-small")
print(converter.smiles_to_iupac('CCO', validate=True))

['ethanol'] 1.0

The larger is Tanimoto similarity, the larger is probability, that the prediction was correct.

You can also process validation manually:

from chemicalconverters import NamesConverter

validation_model = NamesConverter(model_name="knowledgator/IUPAC2SMILES-canonical-base")
print(NamesConverter.validate_iupac(input_sequence='CCO', predicted_sequence='CCO', validation_model=validation_model))

1.0

Bias, Risks, and Limitations

This model has limited accuracy in processing large molecules and currently, doesn't support isomeric and isotopic SMILES.

Training Procedure

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

The model was trained on 100M examples of SMILES-IUPAC pairs with lr=0.0003, batch_size=1024 for 2 epochs.

Evaluation

Citation

Coming soon.

Model Card Authors

Mykhailo Shtopko

Model Card Contact

[email protected]