Translating the InChI: adapting neural machine translation to predict IUPAC names from a chemical identifier

We present a sequence-to-sequence machine learning model for predicting the IUPAC name of a chemical from its standard International Chemical Identifier (InChI). The model uses two stacks of transformers in an encoder-decoder architecture, a setup similar to the neural networks used in state-of-the-art machine translation. Unlike neural machine translation, which usually tokenizes input and output into words or sub-words, our model processes the InChI and predicts the IUPAC name character by character. The model was trained on a dataset of 10 million InChI/IUPAC name pairs freely downloaded from the National Library of Medicine’s online PubChem service. Training took seven days on a Tesla K80 GPU, and the model achieved a test set accuracy of 91%. The model performed particularly well on organics, with the exception of macrocycles, and was comparable to commercial IUPAC name generation software. The predictions were less accurate for inorganic and organometallic compounds. This can be explained by inherent limitations of standard InChI for representing inorganics, as well as low coverage in the training data.

MM-Deacon: Multimodal molecular domain embedding analysis via contrastive learning

Molecular representation learning plays an essential role in cheminformatics. Recently, language model-based approaches have been popular as an alternative to traditional expert-designed features to encode molecules. However, these approaches only utilize a single modality for representing molecules. Driven by the fact that a given molecule can be described through different modalities such as Simplified Molecular Line Entry System (SMILES), The International Union of Pure and Applied Chemistry (IUPAC), and The IUPAC International Chemical Identifier (InChI), we propose a multimodal molecular embedding generation approach called MM-Deacon (multimodal molecular domain embedding analysis via contrastive learning). MM-Deacon is trained using SMILES and IUPAC molecule representations as two different modalities. First, SMILES and IUPAC strings are encoded by using two different transformer-based language models independently, then the contrastive loss is utilized to bring these encoded representations from different modalities closer to each other if they belong to the same molecule, and to push embeddings farther from each other if they belong to different molecules. We evaluate the robustness of our molecule embeddings on molecule clustering, cross-modal molecule search, drug similarity assessment and drug-drug interaction tasks.

InChI version 1.06: now more than 99.99% reliable

The software for the IUPAC Chemical Identifier, InChI, is extraordinarily reliable. It has been tested on large databases around the world, and has proved itself to be an essential tool in the handling and integration of large chemical databases. InChI version 1.05 was released in January 2017 and version 1.06 in December 2020. In this paper, we report on the current state of the InChI Software, the details of the improvements in the v1.06 release, and the results of a test of the InChI run on PubChem, a database of more than a hundred million molecules. The upgrade introduces significant new features, including support for pseudo-element atoms and an improved description of polymers. We expect that few, if any, applications using the standard InChI will need to change as a result of the changes in version 1.06. Numerical instability was discovered for 0.002% of this database, and a small number of other molecules were discovered for which the algorithm did not run smoothly. On the basis of PubChem data, we can demonstrate that InChI version 1.05 was 99.996% accurate, and InChI version 1.06 represents a step closer to perfection. Finally, we look forward to future releases and extensions for the InChI Chemical identifier.

Enhancing the Discovery of Chemistry Theses by Registering Substances and Depositing in PubChem

Chemical substances from theses are not widely accessible as searchable machine-readable formats. In this article, we describe our workflow for extracting, registering, and sharing chemical substances from the University of Alabama theses to enhance discovery. In total, 73 theses were selected for the project, resulting in about 3,000 substances registered using the IUPAC International Chemical Identifier and deposited in PubChem as either structure-data files or Simplified Molecular-Input Line-Entry System notations. In addition to substances being deposited in PubChem, an archive copy was also deposited in the University of Alabama Institutional Repository. The PubChem records for the substance depositions include the full bibliographic reference and link to the thesis full text or thesis metadata when the full text is not yet available. Excluding mixtures, we found that 40% of the shared substances were new to PubChem at the time of deposition. We conclude this article with a detailed discussion about our experiences, challenges, and recommendations for librarians and curators engaged in sharing chemical substance data from theses and similar documents.

InChI Version 1.06: Now More Than 99.99 % Reliable

The software for the IUPAC Chemical Identifier, InChI, is extraordinarily reliable. It has been tested on large databases around the world, and has proved itself to be an essential tool in the handling and integration of large chemical databases. InChI version 1.05 was released in January 2017 and v. 1.06 in December 2020. In this paper, we report on the current state of InChI Software, the details of the improvements in the v.1.06 release , and the results of a test of the InChI run on PubChem, a database of more than a hundred million molecules. The upgrade introduces significant new features, including support for pseudo-element atoms and an improved description of polymers. We expect that few, if any, applications using the standard InChI will need to change as a result of the changes in version 1.06. Numerical instability was discovered for 0.002 % of this database, and a small number of other molecules were discovered for which the algorithm did not run smoothly. On the basis of PubChem data, we can demonstrate that InChI version 1.05 was 99.996 % accurate, and InChI version 1.06 represents a step closer to perfection. Finally, we look forward to future releases and extensions for the InChI Chemical identifier.

Translating the Molecules: Adapting Neural Machine Translation to Predict IUPAC Names from a Chemical Identifier

We present a sequence-to-sequence machine learning model for predicting the IUPAC name of a chemical from its standard International Chemical Identifier (InChI). The model uses two stacks of transformers in an encoder-decoder architecture, a setup similar to the neural networks used in state-of-the-art machine translation. Unlike neural machine translation, which usually tokenizes input and output into words or sub-words, our model processes the InChI and predicts the 2 IUPAC name character by character. The model was trained on a dataset of 10 million InChI/IUPAC name pairs freely downloaded from the National Library of Medicine’s online PubChem service. Training took five days on a Tesla K80 GPU, and the model achieved test-set accuracies of 95% (character-level) and 91% (whole name). The model performed particularly well on organics, with the exception of macrocycles. The predictions were less accurate for inorganic compounds, with a character-level accuracy of 71%. This can be explained by inherent limitations in InChI for representing inorganics, as well as low coverage (1.4 %) of the training data.

Translating the Molecules: Adapting Neural Machine Translation to Predict IUPAC Names from a Chemical Identifier

We present a sequence-to-sequence machine learning model for predicting the IUPAC name of a chemical from its standard International Chemical Identifier (InChI). The model uses two stacks of transformers in an encoder-decoder architecture, a setup similar to the neural networks used in state-of-the-art machine translation. Unlike neural machine translation, which usually tokenizes input and output into words or sub-words, our model processes the InChI and predicts the 2 IUPAC name character by character. The model was trained on a dataset of 10 million InChI/IUPAC name pairs freely downloaded from the National Library of Medicine’s online PubChem service. Training took five days on a Tesla K80 GPU, and the model achieved test-set accuracies of 95% (character-level) and 91% (whole name). The model performed particularly well on organics, with the exception of macrocycles. The predictions were less accurate for inorganic compounds, with a character-level accuracy of 71%. This can be explained by inherent limitations in InChI for representing inorganics, as well as low coverage (1.4 %) of the training data.

Can an InChI for Nano Address the Need for a Simplified Representation of Complex Nanomaterials across Experimental and Nanoinformatics Studies?

Chemoinformatics has developed efficient ways of representing chemical structures for small molecules as simple text strings, simplified molecular-input line-entry system (SMILES) and the IUPAC International Chemical Identifier (InChI), which are machine-readable. In particular, InChIs have been extended to encode formalized representations of mixtures and reactions, and work is ongoing to represent polymers and other macromolecules in this way. The next frontier is encoding the multi-component structures of nanomaterials (NMs) in a machine-readable format to enable linking of datasets for nanoinformatics and regulatory applications. A workshop organized by the H2020 research infrastructure NanoCommons and the nanoinformatics project NanoSolveIT analyzed issues involved in developing an InChI for NMs (NInChI). The layers needed to capture NM structures include but are not limited to: core composition (possibly multi-layered); surface topography; surface coatings or functionalization; doping with other chemicals; and representation of impurities. NM distributions (size, shape, composition, surface properties, etc.), types of chemical linkages connecting surface functionalization and coating molecules to the core, and various crystallographic forms exhibited by NMs also need to be considered. Six case studies were conducted to elucidate requirements for unambiguous description of NMs. The suggested NInChI layers are intended to stimulate further analysis that will lead to the first version of a “nano” extension to the InChI standard.