Learning continuous and data-driven molecular descriptors by translating equivalent chemical representations

By R. Winter, F. Montanari, Frank No'e, Djork-Arné Clevert. 2018

390 citations recorded by the source corpus.

Translation between semantically equivalent but syntactically different line notations of molecular structures compresses meaningful information into a continuous molecular descriptor.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Spectroscopy

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Materials Chemistry

In Spectroscopy

Citations in the corpus

Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties (2020, background): “1 Related Work Molecular properties, which depend only on the atomic distance graph, are commonly predicted by kernel methods or Gaussian process regression [14, 15, 16, 17] or graph neural networks [18, 19], where ENNs are usually employed for predicting physical properties, which depend on atomic ”
Uni-Mol: A Universal 3D Molecular Representation Learning Framework (2023, methodology): “In most current MRL methods, one starts with representing molecules as 1D sequential strings, such as SMILES [16, 17, 18] and InChI [19, 20, 21], or 2D graphs [22, 11, 23, 12, 24].”
SMILES Transformer: Pre-trained Molecular Fingerprint for Low Data Drug Discovery (2019, methodology): “…Vinyals, and Le 2014) or variational autoencoders (VAE) (Kingma and Welling 2014) by decoding SMILES from the learned representations (G´omez-Bombarelli et al. 2018; Xu et al. 2017; Kusner, Paige, and Hern´andez-Lobato 2017; Goh et al. 2018; Bjerrum and Sattarov 2018; Winter et al. 2019).”
Generating valid Euclidean distance matrices (2019, background): “, the molecular graph structure [7, 8, 9], and based on the geometry, i.”
Machine learning for molecular simulation (2019, methodology): “These methods have been used to draw photorealistic images [92, 93], generate speech or music audio [94] and generate chemical structures to design molecules or materials with certain properties [42, 43, 44, 45].”
All SMILES Variational Autoencoder (2019, methodology): “To accommodate this intentionally difﬁcult reconstruction task, facilitate the construction of a bijection between latent space and molecules, and following prior work [28, 67], we use a width-5 beam search decoder to map from the latent representation to the space of molecules at test-time.”
ChemBERTa-2: Towards Chemical Foundation Models (2022, background): “Other work attempts to combine molecular graph and transformer based pretraining methodologies into a uniﬁed "dual" framework [14], or considers techniques inspired by neural machine translation by learning to translate between SMILES and InChi representations of a molecule [15].”
Mol-CycleGAN: a generative model for molecular optimization (2019, background): “One of the main approaches is de novo design, that is using generative models to propose new molecules that are likely to possess the desired properties [3,5,15,17].”
De novo generation of hit-like molecules from gene expression signatures using artificial intelligence (2018, methodology): “In this sense, the loss functions for 𝐺 1 (𝑠 0 , 𝑐) and 𝐷 1 (𝑥) can be written as: All molecular structures were encoded into a vector of continuous values using an approach similar to the one developed by Winter et al. 40 based on molecular translation.”
Artificial Intelligence For Natural Product Drug Discovery (2023, methodology): “Moreover, encoder–decoder architectures are used to featurize compounds for virtual screening from different input formats 120–122 .”