In the past few years, there has been considerable activity in both academic and industrial research to develop innovative machine learning approaches to locate novel, high-performing molecules in chemical space. Here we describe a new and fundamentally different type of approach that provides a holistic overview of how high-performing molecules are distributed throughout a search space. Based on an open-source, graph-based implementation [J. H. Jensen, , 2019, , 3567-3572] of a traditional genetic algorithm for molecular optimisation, and influenced by state-of-the-art concepts from soft robot design [J. B. Mouret and J. Clune, , 2012, pp. 593-594], we provide an algorithm that (i) produces a large diversity of high-performing, yet qualitatively different molecules, (ii) illuminates the distribution of optimal solutions, and (iii) improves search efficiency compared to both machine learning and traditional genetic algorithm approaches.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8162856PMC
http://dx.doi.org/10.1039/d0sc03544kDOI Listing

Publication Analysis

Top Keywords

chemical space
8
machine learning
8
high-performing molecules
8
traditional genetic
8
genetic algorithm
8
illuminating elite
4
elite patches
4
patches chemical
4
space years
4
years considerable
4

Similar Publications

Meteoritic impacts on planetary surfaces deliver a significant amount of energy that can produce prebiotic organic compounds such as cyanides, which may be a key step to the formation of biomolecules. To study the chemical processes of impact-induced organic synthesis, we simulated the physicochemical processes of hypervelocity impacts (HVI) in experiments with both high-speed C projectiles and laser ablation. In the first approach, a C beam was accelerated to collide with ammonium nitrate (NHNO) to reproduce the shock process and plume generation of meteoritic impacts on nitrogen-rich planetary surfaces.

View Article and Find Full Text PDF

Waste Heat and Habitability: Constraints from Technological Energy Consumption.

Astrobiology

January 2025

Department of Aerospace, Physics and Space Sciences, Florida Institute of Technology, Melbourne, Florida, USA.

Waste heat production represents an inevitable consequence of energy conversion as per the laws of thermodynamics. Based on this fact, by using simple theoretical models, we analyze constraints on the habitability of Earth-like terrestrial planets hosting putative technological species and technospheres characterized by persistent exponential growth of energy consumption and waste heat generation. In particular, we quantify the deleterious effects of rising surface temperature on biospheric processes and the eventual loss of liquid water.

View Article and Find Full Text PDF

Peripheral Evolution of Tanshinone IIA and Cryptotanshinone for Discovery of a Potent and Specific NLRP3 Inflammasome Inhibitor.

J Med Chem

January 2025

College of Pharmaceutical Sciences, State Key Laboratory of Advanced Drug Delivery and Release Systems, Zhejiang University, Hangzhou 310058, China.

Natural products (NPs) continue to serve as an invaluable source in drug discovery, and peripheral evolution of NPs is a highly efficient evolution strategy. Herein, we describe a unified "methyl to amide" peripheral evolution of Tanshinone IIA and Cryptotanshinone for discovery of NLRP3 inflammasome inhibitors. There were 54 compounds designed and prepared, while the chemoinformatic analysis revealed that these evolved NP analogues occupy a unique chemical space.

View Article and Find Full Text PDF

Identifying facile strategies for hierarchically structuring crystalline porous materials is critical for realizing diffusion length scales suitable for broad applications. Here, we elucidate synthesis-structure-function relations governing how room temperature catalytic conditions can be exploited to tune covalent organic framework (COF) growth and thereby access unique hierarchical morphologies without the need to introduce secondary templates or structure directing molecules. Specifically, we demonstrate how scandium triflate, an efficient catalyst involved in the synthesis of imine-based COFs, can be exploited as an effective growth modifier capable of selectively titrating terminal amines on 2D COF layers to facilitate anisotropic crystal growth.

View Article and Find Full Text PDF

Crystal Growth and Structural Analysis of Layered Lithium Titanium Sulfide.

Inorg Chem

January 2025

Department of Materials Chemistry, Faculty of Engineering, Shinshu University, 4-17-1 Wakasato, Nagano 380-8553, Japan.

Layered sulfide crystals are suitable hosts for lithium and sodium ions in batteries. In this study, new layered lithium titanium sulfide (LTS) crystals were grown in a sealed silica tube using a LiS self-flux at 800-950 °C. X-ray diffraction (XRD) analysis results indicated the formation of a new sulfide phase with higher symmetry in the Li-Ti-S system.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!