•Journal•ISSN: 2374-7943
ACS central science
American Chemical Society
About: ACS central science is an academic journal published by American Chemical Society. The journal publishes majorly in the area(s): Medicine & Chemistry. It has an ISSN identifier of 2374-7943. It is also open access. Over the lifetime, 1657 publications have been published receiving 77739 citations. The journal is also known as: American Chemical Society central science & ACS Cent Sci.
Topics: Medicine, Chemistry, Computer science, Biology, Catalysis
Papers
More filters
TL;DR: In this article, a deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder, and a predictor, which can generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds.
Abstract: We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation. This model allows us to generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds. A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder, and a predictor. The encoder converts the discrete representation of a molecule into a real-valued continuous vector, and the decoder converts these continuous vectors back to discrete molecular representations. The predictor estimates chemical properties from the latent continuous vector representation of the molecule. Continuous representations of molecules allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules. Continuous represent...
1,884 citations
TL;DR: This work shows that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing, and demonstrates that the properties of the generated molecules correlate very well with those of the molecules used to train the model.
Abstract: In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active toward a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria), it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecule...
1,041 citations
TL;DR: Since the outbreak of the novel coronavirus disease COVID-19, caused by the SARS-CoV-2 virus, this disease has spread rapidly around the globe and the potential threat of a pandemic is considered.
Abstract: Since the outbreak of the novel coronavirus disease COVID-19, caused by the SARS-CoV-2 virus, this disease has spread rapidly around the globe. Considering the potential threat of a pandemic, scien...
1,029 citations
TL;DR: An outlook on lithium ion technology is presented by providing first the current status and then the progress and challenges with the ongoing approaches, and finally points out practically viable near-term strategies.
Abstract: Lithium ion batteries as a power source are dominating in portable electronics, penetrating the electric vehicle market, and on the verge of entering the utility market for grid-energy storage. Depending on the application, trade-offs among the various performance parameters—energy, power, cycle life, cost, safety, and environmental impact—are often needed, which are linked to severe materials chemistry challenges. The current lithium ion battery technology is based on insertion-reaction electrodes and organic liquid electrolytes. With an aim to increase the energy density or optimize the other performance parameters, new electrode materials based on both insertion reaction and dominantly conversion reaction along with solid electrolytes and lithium metal anode are being intensively pursued. This article presents an outlook on lithium ion technology by providing first the current status and then the progress and challenges with the ongoing approaches. In light of the formidable challenges with some of the...
814 citations
TL;DR: An ongoing theme of the COVID-19 pandemic is the need for widespread availability of accurate and efficient diagnostic testing for detection of SARS-CoV-2 and antiviral antibodies in infected indiv...
Abstract: An ongoing theme of the COVID-19 pandemic is the need for widespread availability of accurate and efficient diagnostic testing for detection of SARS-CoV-2 and antiviral antibodies in infected indiv...
732 citations