![](\"https://media.springernature.com/lw685/springer-static/image/art%3A10.1186%2F1758-2946-5-14/MediaObjects/13321_2012_Article_463_Fig2_HTML.jpg\")
Automatic interpretation of chemical compounds images, to convert the image in structured machine-readable chemical descriptions (InChI) is of paramount importance, with a potential to speed-up tremendously the research in key areas, like chemistry, bioengineering, drug design and drug testing. In the following, we review a series of papers and other available resources for this research domain.
\n
A comprehensive review of literature about optical chemical structure recognition tools - from as early as 1993 to present days
\nA review of optical chemical structure recognition tools
Transformer-based models that predict SMILES strings from chemical compound names Transformer-based Approach for Predicting Chemical Compound Structures
Transformer-CNN method using SMILES augmentation for training and inference
\nTransformer-CNN: Swiss knife for QSAR modeling and interpretation
Model based on Deep Learning to predict the graph structure of the molecule and produce all information necessary to relate each component of the resulting graph to the source image
\nChemGrapher: Optical Graph Recognition of Chemical Compounds by Deep Learning
End-to-end deep learning solutions for both segmenting molecular structures from documents and for predicting chemical structures from these segmented images
\nMolecular Structure Extraction From Documents Using Deep Learning
Generative algorithms for new molecules
\n3DMolNet: A Generative Network for Molecular Structures
Chemical-Chemical Interaction prediction using Deep Learning
\nDeepCCI: End-to-end Deep Learning for Chemical-Chemical Interaction Prediction
A review that outlines the impact made by advances in NLP on drug discovery and aims to further the dialogue between medicinal chemists and computer scientists.
\nExploring Chemical Space using Natural Language Processing Methodologies for Drug Discovery
SMILES - stands for simplified molecular-input line-entry system (SMILES) is a specification in the form of a line notation for describing the structure of chemical species using short ASCII strings. SMILES strings can be imported by most molecule editors for conversion back into two-dimensional drawings or three-dimensional models of the molecules.
\nSELFIES stands for SELF-referencIng Embedding Strings (SELFIES), an alternative sequence-based representation that is built upon “semantically constrained graphs”.
\nSLN - is a line notation also known as SYBYL, a specification for unambiguously describing the structure of chemical molecules using short ASCII strings. SLN can specify molecules, molecular queries, and reactions in a single line notation whereas SMILES handles these through language extensions.
\nWSN - Wiswesser line notation was the first line notation capable of precisely describing complex molecules. WLN allowed for indexing the Chemical Structure Index (CSI) at the Institute for Scientific Information (ISI).
\nInChl - is the IUPAC International Chemical Identifier, which is a nonproprietary and open-source structural representation (inchi-trust.org).
", "messages": [ { "id": "1226543", "postDate": "03/04/2021 16:16:29", "content": "Automatic interpretation of chemical compounds images, to convert the image in structured machine-readable chemical descriptions (InChI) is of paramount importance, with a potential to speed-up tremendously the research in key areas, like chemistry, bioengineering, drug design and drug testing. In the following, we review a series of papers and other available resources for this research domain.
\n
A comprehensive review of literature about optical chemical structure recognition tools - from as early as 1993 to present days
\nA review of optical chemical structure recognition tools
Transformer-based models that predict SMILES strings from chemical compound names Transformer-based Approach for Predicting Chemical Compound Structures
Transformer-CNN method using SMILES augmentation for training and inference
\nTransformer-CNN: Swiss knife for QSAR modeling and interpretation
Model based on Deep Learning to predict the graph structure of the molecule and produce all information necessary to relate each component of the resulting graph to the source image
\nChemGrapher: Optical Graph Recognition of Chemical Compounds by Deep Learning
End-to-end deep learning solutions for both segmenting molecular structures from documents and for predicting chemical structures from these segmented images
\nMolecular Structure Extraction From Documents Using Deep Learning
Generative algorithms for new molecules
\n3DMolNet: A Generative Network for Molecular Structures
Chemical-Chemical Interaction prediction using Deep Learning
\nDeepCCI: End-to-end Deep Learning for Chemical-Chemical Interaction Prediction
A review that outlines the impact made by advances in NLP on drug discovery and aims to further the dialogue between medicinal chemists and computer scientists.
\nExploring Chemical Space using Natural Language Processing Methodologies for Drug Discovery
SMILES - stands for simplified molecular-input line-entry system (SMILES) is a specification in the form of a line notation for describing the structure of chemical species using short ASCII strings. SMILES strings can be imported by most molecule editors for conversion back into two-dimensional drawings or three-dimensional models of the molecules.
\nSELFIES stands for SELF-referencIng Embedding Strings (SELFIES), an alternative sequence-based representation that is built upon “semantically constrained graphs”.
\nSLN - is a line notation also known as SYBYL, a specification for unambiguously describing the structure of chemical molecules using short ASCII strings. SLN can specify molecules, molecular queries, and reactions in a single line notation whereas SMILES handles these through language extensions.
\nWSN - Wiswesser line notation was the first line notation capable of precisely describing complex molecules. WLN allowed for indexing the Chemical Structure Index (CSI) at the Institute for Scientific Information (ISI).
\nInChl - is the IUPAC International Chemical Identifier, which is a nonproprietary and open-source structural representation (inchi-trust.org).
", "rawMarkdown": "### Introduction\n\nAutomatic interpretation of chemical compounds images, to convert the image in structured machine-readable chemical descriptions (InChI) is of paramount importance, with a potential to speed-up tremendously the research in key areas, like chemistry, bioengineering, drug design and drug testing. In the following, we review a series of papers and other available resources for this research domain.\n\n\n\n### Papers\n\n* A comprehensive review of literature about optical chemical structure recognition tools - from as early as 1993 to present days \n[A review of optical chemical structure recognition tools](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-020-00465-0) \n\n* Transformer-based models that predict SMILES strings from chemical compound names [Transformer-based Approach for Predicting Chemical Compound Structures](https://www.aclweb.org/anthology/2020.aacl-main.19.pdf) \n\n* Transformer-CNN method using SMILES augmentation for training and inference\n[Transformer-CNN: Swiss knife for QSAR modeling and interpretation](https://link.springer.com/article/10.1186/s13321-020-00423-w)\n\n* Model based on Deep Learning to predict the graph structure of the molecule and produce all information necessary to relate each component of the resulting graph to the source image\n [ChemGrapher: Optical Graph Recognition of Chemical Compounds by Deep Learning](https://arxiv.org/pdf/2002.09914.pdf) \n\n* End-to-end deep learning solutions for both segmenting molecular structures from documents and for predicting chemical structures from these segmented images\n [Molecular Structure Extraction From Documents Using Deep Learning](https://arxiv.org/ftp/arxiv/papers/1802/1802.04903.pdf) \n\n* Generative algorithms for new molecules \n[3DMolNet: A Generative Network for Molecular Structures](https://arxiv.org/pdf/2010.06477.pdf) \n\n* Chemical-Chemical Interaction prediction using Deep Learning\n[DeepCCI: End-to-end Deep Learning for Chemical-Chemical Interaction Prediction](https://arxiv.org/pdf/1704.08432.pdf)\n\n* A review that outlines the impact made by advances in NLP on drug discovery and aims to further the dialogue between medicinal chemists and computer scientists.\n [Exploring Chemical Space using Natural Language Processing Methodologies for Drug Discovery](https://arxiv.org/pdf/2002.06053.pdf) \n\n\n### Projects\n\n* SMILES Transformer - extracts molecular fingerprints from string representations of chemical molecules: https://github.com/DSPsleeporg/smiles-transformer\n\n### Glossary\n\n**SMILES** - stands for simplified molecular-input line-entry system (SMILES) is a specification in the form of a line notation for describing the structure of chemical species using short ASCII strings. SMILES strings can be imported by most molecule editors for conversion back into two-dimensional drawings or three-dimensional models of the molecules. \n\n**SELFIES** stands for SELF-referencIng Embedding Strings (SELFIES), an alternative sequence-based representation that is built upon “semantically constrained graphs”. \n\n**SLN** - is a line notation also known as SYBYL, a specification for unambiguously describing the structure of chemical molecules using short ASCII strings. SLN can specify molecules, molecular queries, and reactions in a single line notation whereas SMILES handles these through language extensions. \n\n**WSN** - Wiswesser line notation was the first line notation capable of precisely describing complex molecules. WLN allowed for indexing the Chemical Structure Index (CSI) at the Institute for Scientific Information (ISI). \n\n**InChl** - is the IUPAC International Chemical Identifier, which is a nonproprietary and open-source structural representation (inchi-trust.org).",
"votes": null
},
{
"id": "1226855",
"postDate": "03/05/2021 00:09:02",
"content": "Thanks for share these papers!
", "rawMarkdown": "Thanks for share these papers!", "votes": null }, { "id": "1230339", "postDate": "03/08/2021 03:23:07", "content": "@gpreda Thanks for sharing Gabriel . Thats quite an extensive list of papers , glossary and projects . SMILES Transformer is particularly insightful
", "rawMarkdown": "gpreda Thanks for sharing Gabriel . Thats quite an extensive list of papers , glossary and projects . SMILES Transformer is particularly insightful", "votes": null } ], "comments": [ { "id": 1226855, "author_name": "yuyafukiharu", "author_url": "", "post_date": "03/05/2021 00:09:02", "content": "Thanks for share these papers!
", "votes": null, "replies": [] }, { "id": 1230339, "author_name": "usharengaraju", "author_url": "", "post_date": "03/08/2021 03:23:07", "content": "@gpreda Thanks for sharing Gabriel . Thats quite an extensive list of papers , glossary and projects . SMILES Transformer is particularly insightful
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1226543": "### Introduction\n\nAutomatic interpretation of chemical compounds images, to convert the image in structured machine-readable chemical descriptions (InChI) is of paramount importance, with a potential to speed-up tremendously the research in key areas, like chemistry, bioengineering, drug design and drug testing. In the following, we review a series of papers and other available resources for this research domain.\n\n\n\n### Papers\n\n* A comprehensive review of literature about optical chemical structure recognition tools - from as early as 1993 to present days \n[A review of optical chemical structure recognition tools](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-020-00465-0) \n\n* Transformer-based models that predict SMILES strings from chemical compound names [Transformer-based Approach for Predicting Chemical Compound Structures](https://www.aclweb.org/anthology/2020.aacl-main.19.pdf) \n\n* Transformer-CNN method using SMILES augmentation for training and inference\n[Transformer-CNN: Swiss knife for QSAR modeling and interpretation](https://link.springer.com/article/10.1186/s13321-020-00423-w)\n\n* Model based on Deep Learning to predict the graph structure of the molecule and produce all information necessary to relate each component of the resulting graph to the source image\n [ChemGrapher: Optical Graph Recognition of Chemical Compounds by Deep Learning](https://arxiv.org/pdf/2002.09914.pdf) \n\n* End-to-end deep learning solutions for both segmenting molecular structures from documents and for predicting chemical structures from these segmented images\n [Molecular Structure Extraction From Documents Using Deep Learning](https://arxiv.org/ftp/arxiv/papers/1802/1802.04903.pdf) \n\n* Generative algorithms for new molecules \n[3DMolNet: A Generative Network for Molecular Structures](https://arxiv.org/pdf/2010.06477.pdf) \n\n* Chemical-Chemical Interaction prediction using Deep Learning\n[DeepCCI: End-to-end Deep Learning for Chemical-Chemical Interaction Prediction](https://arxiv.org/pdf/1704.08432.pdf)\n\n* A review that outlines the impact made by advances in NLP on drug discovery and aims to further the dialogue between medicinal chemists and computer scientists.\n [Exploring Chemical Space using Natural Language Processing Methodologies for Drug Discovery](https://arxiv.org/pdf/2002.06053.pdf) \n\n\n### Projects\n\n* SMILES Transformer - extracts molecular fingerprints from string representations of chemical molecules: https://github.com/DSPsleeporg/smiles-transformer\n\n### Glossary\n\n**SMILES** - stands for simplified molecular-input line-entry system (SMILES) is a specification in the form of a line notation for describing the structure of chemical species using short ASCII strings. SMILES strings can be imported by most molecule editors for conversion back into two-dimensional drawings or three-dimensional models of the molecules. \n\n**SELFIES** stands for SELF-referencIng Embedding Strings (SELFIES), an alternative sequence-based representation that is built upon “semantically constrained graphs”. \n\n**SLN** - is a line notation also known as SYBYL, a specification for unambiguously describing the structure of chemical molecules using short ASCII strings. SLN can specify molecules, molecular queries, and reactions in a single line notation whereas SMILES handles these through language extensions. \n\n**WSN** - Wiswesser line notation was the first line notation capable of precisely describing complex molecules. WLN allowed for indexing the Chemical Structure Index (CSI) at the Institute for Scientific Information (ISI). \n\n**InChl** - is the IUPAC International Chemical Identifier, which is a nonproprietary and open-source structural representation (inchi-trust.org).",
"1226855": "Thanks for share these papers!",
"1230339": "gpreda Thanks for sharing Gabriel . Thats quite an extensive list of papers , glossary and projects . SMILES Transformer is particularly insightful"
},
"source": "meta"
}