The chemical space project.

By Jean-Louis Reymond. 2015

596 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Environmental Chemistry, Physical and Theoretical Chemistry, Computational Chemistry & Drug Discovery

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 Environmental Chemistry

In Physical and Theoretical Chemistry

In Computational Chemistry & Drug Discovery

Citations in the corpus

SchNet: A continuous-filter convolutional neural network for modeling quantum interactions (2017, background): “QM9 is a widely used benchmark for the prediction of various molecular properties in equilibrium [8, 38, 39].”
Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements (2021, result): “This is a typical case in which NNP is able to achieve the following three properties at the same time: (1) the ability to handle a wide variety of elements, (2) the ability to handle phenomena that were not assumed at the time of training, and (3) a significantly faster speed than that of DFT.”
Randomized SMILES strings improve the quality of molecular generative models (2019, background): “One of the most interesting attempts to understand the chemical space is the GDB project [2], which encompasses a set of databases that combinatorially enumerate large parts of the small molecule fragment-like chemical space.”
Optimizing distributions over molecular space. An Objective-Reinforced Generative Adversarial Network for Inverse-design Chemistry (ORGANIC) (2017, background): “Current strategies to explore and discover molecules lie with high-throughput virtual screening [2] or discrete local search methods such as genetic algorithms [3–5].”
Deep reinforcement learning for de novo drug design (2017, background): “nal chemists face virtually infinite chemical space. The diversity of synthetically feasible chemicals that can be considered as potential drug-like molecules is estimated to be between 1030 and 1060 (22, 23). Great advances in both computational algorithms(24, 25), hardware, and high-throughput scr”
Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models (2018, background): “It has been estimated that the number of pharmacologically-sensible molecules is in the order of 10 23 to 10 80 compounds [3, 4].”
Large Language Models on Graphs: A Comprehensive Survey (2023, background): “Chemical space has been found to contain more than 10 60 molecules [198], which is beyond the capacity of exploration in wet lab experiments.”
Artificial Intelligence For Natural Product Drug Discovery (2023, background): “Chemical space — typically defined by using multiple compound properties of interest, such as physicochemical properties — is vast and largely unexplored 224 .”