Search results

Search for "deep learning" in Full Text gives 13 result(s) in Beilstein Journal of Organic Chemistry.

Emerging trends in the optimization of organic synthesis through high-throughput tools and machine learning

  • Pablo Quijano Velasco,
  • Kedar Hippalgaonkar and
  • Balamurugan Ramalingam

Beilstein J. Org. Chem. 2025, 21, 10–38, doi:10.3762/bjoc.21.3

Graphical Abstract
PDF
Album
Review
Published 06 Jan 2025

Machine learning-guided strategies for reaction conditions design and optimization

  • Lung-Yi Chen and
  • Yi-Pei Li

Beilstein J. Org. Chem. 2024, 20, 2476–2492, doi:10.3762/bjoc.20.212

Graphical Abstract
  • of natural language processing (NLP) techniques to extract experimental data from unstructured text. For example, Vaucher et al. [69] combined rule-based models and deep-learning techniques to convert experimental procedures into standardized synthetic steps. They further used this data extraction
  • ) text-based featurization, as shown in Figure 2. Descriptor-based methods are often used for datasets with limited samples, since they incorporate chemistry- or physics-informed features that can enhance the model's ability to fit the data. Graph-based and text-based methods rely on deep-learning
  • . Therefore, a preprocessing step to standardize the labels and reduce redundancy is also essential. Gao et al. [18] developed a large-scale model for predicting reaction conditions, using a deep learning approach trained on the Reaxys database. Their model could sequentially predict the catalysts, solvents
PDF
Album
Review
Published 04 Oct 2024

Catalysing (organo-)catalysis: Trends in the application of machine learning to enantioselective organocatalysis

  • Stefan P. Schmid,
  • Leon Schlosser,
  • Frank Glorius and
  • Kjell Jorner

Beilstein J. Org. Chem. 2024, 20, 2280–2304, doi:10.3762/bjoc.20.196

Graphical Abstract
  • addition, such larger data sets also lead to an increased interest in the application of deep learning tools, such as graph-based neural networks, to organocatalysis. One particular example was published by Hong and co-workers [113], who developed a chemistry-informed graph model for the prediction of
PDF
Album
Review
Published 10 Sep 2024

Finding the most potent compounds using active learning on molecular pairs

  • Zachary Fralish and
  • Daniel Reker

Beilstein J. Org. Chem. 2024, 20, 2152–2162, doi:10.3762/bjoc.20.185

Graphical Abstract
  • default parameters and aggregation = ‘sum’ using the PyTorch deep learning framework. For the single-molecule Chemprop implementation, number_of_molecules = 1 while for the ActiveDelta implementation number_of_molecules = 2 to allow for processing of multiple inputs as described previously [29]. We
  • multiple lead series to enable further development of distinct scaffolds, but this approach also enriches the scaffold diversity of “negative” training data to improve future compound selection. Although the deep learning-based ActiveDelta models were not able to identify a larger number of hit compounds
  • than the tree-based ActiveDelta implementations here, a deep learning approach appears to be more advantageous to identify more diverse hits by selecting a greater number of distinct scaffolds during exploitative active learning. Analyzing chemical trajectories We next investigated how these models
PDF
Album
Supp Info
Full Research Paper
Published 27 Aug 2024

Computational toolbox for the analysis of protein–glycan interactions

  • Ferran Nieto-Fabregat,
  • Maria Pia Lenza,
  • Angela Marseglia,
  • Cristina Di Carluccio,
  • Antonio Molinaro,
  • Alba Silipo and
  • Roberta Marchetti

Beilstein J. Org. Chem. 2024, 20, 2084–2107, doi:10.3762/bjoc.20.180

Graphical Abstract
  • has been made more accessible. The same research group also developed RoseTTAFold, which uses deep learning to quickly and accurately predict protein structures based on limited information [92]. However, very accurate structures for complex proteins are yet to be achieved at a level suitable for
  • computational tools have been developed to facilitate the prediction of protein binding sites. We report here only the applications related to the protein interaction with glycans: 1. PeSTo-Carbs [105]: it is an extension of Protein Structure Transformer (PeSTo) [106], a deep learning method to predict protein
  • range of carbohydrates, their derivatives and cyclodextrins, and a specific model PS-S for important carbohydrate monomers. All of these features are available for free without registration as online tools (https://pesto.epfl.ch/). 2. GlyNet [107]: it is a free deep learning algorithm, based on neural
PDF
Album
Review
Published 22 Aug 2024

Hetero-polycyclic aromatic systems: A data-driven investigation of structure–property relationships

  • Sabyasachi Chakraborty,
  • Eduardo Mayo Yanes and
  • Renana Gershoni-Poranne

Beilstein J. Org. Chem. 2024, 20, 1817–1830, doi:10.3762/bjoc.20.160

Graphical Abstract
  • PASs and their molecular properties. The first installment, COMPAS-1 [6], contains ~35k cata-condensed polybenzenoid hydrocarbons (cc-PBHs) and has already enabled various directions of investigation, including by training of both interpretable machine [7] and deep learning methods [8], which led to
  • relationships require more advanced data-analysis tools, and we are currently leveraging different machine learning and deep learning techniques to tap the full potential of the COMPAS-2 dataset. A) The building blocks used in the COMPAS-2 datasets. B) Possible annulation types formed when combining the
PDF
Album
Supp Info
Full Research Paper
Published 31 Jul 2024

pKalculator: A pKa predictor for C–H bonds

  • Rasmus M. Borup,
  • Nicolai Ree and
  • Jan H. Jensen

Beilstein J. Org. Chem. 2024, 20, 1614–1622, doi:10.3762/bjoc.20.144

Graphical Abstract
  • database only contains an image of each molecule, we employ the “Deep Learning for Chemical Image Recognition” software (DECIMER v. 2.0), developed by Rajan and co-workers [8][9][10]. While DECIMER converts molecular images into SMILES, manual intervention is required to ensure the SMILES string correctly
PDF
Album
Supp Info
Full Research Paper
Published 16 Jul 2024

Chemoenzymatic synthesis of macrocyclic peptides and polyketides via thioesterase-catalyzed macrocyclization

  • Senze Qiao,
  • Zhongyu Cheng and
  • Fuzhuo Li

Beilstein J. Org. Chem. 2024, 20, 721–733, doi:10.3762/bjoc.20.66

Graphical Abstract
  • stability, etc. Emerging research methods on bioinformatics, computational modeling, deep learning, protein engineering, and high-throughput screening will accelerate the pace of enzyme discovery to provide a broader platform of tools for employing chemoenzymatic strategies [64][87][88][89]. More
PDF
Album
Review
Published 04 Apr 2024

GlAIcomics: a deep neural network classifier for spectroscopy-augmented mass spectrometric glycans data

  • Thomas Barillot,
  • Baptiste Schindler,
  • Baptiste Moge,
  • Elisa Fadda,
  • Franck Lépine and
  • Isabelle Compagnon

Beilstein J. Org. Chem. 2023, 19, 1825–1831, doi:10.3762/bjoc.19.134

Graphical Abstract
  • intelligence in combination with spectroscopy-augmented mass spectrometry for carbohydrates sequencing and glycomics applications. Keywords: Bayesian neural network; deep learning; glycomics; IR; spectroscopy; Introduction DNA and protein sequencing technologies that aim at determining the structure of a
PDF
Album
Supp Info
Full Research Paper
Published 05 Dec 2023

Navigating and expanding the roadmap of natural product genome mining tools

  • Friederike Biermann,
  • Sebastian L. Wenski and
  • Eric J. N. Helfrich

Beilstein J. Org. Chem. 2022, 18, 1656–1671, doi:10.3762/bjoc.18.178

Graphical Abstract
  • -like structures and prioritized based on the taxonomic distribution of the cluster. decRiPPter was successfully used for the identification of a new lanthipeptide subfamily, providing experimental validation of the algorithm [65]. A more advanced form of supervised learning is deep learning (Figure 4
  • ). An example of a deep learning architecture is the artificial neural network inspired by the human brain architecture. It consists of artificial neurons processing information organized in different layers and connected by synapses [73]. These advanced algorithms often provide higher accuracy in their
PDF
Album
Perspective
Published 06 Dec 2022

On drug discovery against infectious diseases and academic medicinal chemistry contributions

  • Yves L. Janin

Beilstein J. Org. Chem. 2022, 18, 1355–1378, doi:10.3762/bjoc.18.141

Graphical Abstract
  • issue in this regard as a major portion of published data will have to be filtered out before such methods starts to make some tangible headways [57]. For instance, a recent “deep-learning” search for new antibiotics came out with the finding that halicin (1) depicted in Figure 1 was, as many nitro
PDF
Album
Perspective
Published 29 Sep 2022

Models of necessity

  • Timothy Clark and
  • Martin G. Hicks

Beilstein J. Org. Chem. 2020, 16, 1649–1661, doi:10.3762/bjoc.16.137

Graphical Abstract
  • changed recently is the amount of data that is available and the upsurge of deep-learning algorithms, which date back to the late 1960’s [65] but were preceded in chemistry by far simpler back-propagation neural nets [66] and made their first impact around 2015 [67]. At the second Beilstein Bozen
PDF
Album
Commentary
Published 13 Jul 2020

Biomimetic molecular design tools that learn, evolve, and adapt

  • David A Winkler

Beilstein J. Org. Chem. 2017, 13, 1288–1302, doi:10.3762/bjoc.13.125

Graphical Abstract
  • methods and their potential impacts in chemistry, engineering, and medicine. Keywords: automated chemical synthesis; deep learning; evolutionary algorithms; in silico evolution; machine learning; materials design and development; neural networks; Introduction There is still not a clear understanding of
  • future impact. It introduces the most common type of algorithm, machine learning. A discussion of a very useful machine-learning algorithm, the neural network follows, and problems that often arise in their use, and solutions to these difficulties described. A new type of deep learning neural network
  • paradigm shifting new variants called deep learning. We provide a brief summary of these types of machine learning algorithms to assist those organic chemists who are not familiar with them. Traditional backpropagation algorithm A common machine learning algorithm is the backpropagation neural network
PDF
Album
Review
Published 29 Jun 2017
Other Beilstein-Institut Open Science Activities