Search

The integration of organic and enzymatic synthesis enhances molecule construction efficiency. Here, the authors present ChemEnzyRetroPlanner, an AI-driven platform for automated hybrid synthesis planning, improving synthesis route efficiency and sustainability.

Genomic analyses of 2,191 global common wheat accessions and over 47,000 yellow rust (YR) response data points across several environments and pathogen races provide a genome-wide landscape of YR resistance genes and effective alleles.

This meta-analysis finds that planting nitrogen-fixing species increases soil carbon compared with non-nitrogen-fixing species. Using nitrogen-fixing species for forestation, agriculture and land regeneration might therefore offer a nature-based solution to mitigate climate change.

Adverse drug reactions (ADRs) are a severe healthcare concern worldwide, however, accurate prediction of pediatric-specific ADRs remains challenging. Here, the authors present a machine learning approach integrating consensus-driven signal detection and multi-level biological features, significantly enhancing pediatric ADR prediction.

Gut bacteriome dysbiosis has been implicated in the pathogenesis of inflammatory bowel disease (IBD). Here, they authors characterize the small bowel (terminal ileum) virome and bacteriome of patients with Crohn´s Disease (CD), and show that ileal virions from CD patients causally exacerbate intestinal inflammation in IBD mouse models.

Generative models face challenges with low synthesizability of generated molecules. Here, authors develop ClickGen, a deep learning model using modular reactions and reinforcement learning to generate highly diverse, novel, and synthesizable molecules

Colonoscopy is the most commonly used tool to screen for colorectal cancer (CRC). Here, the authors develop a deep learning model to perform optical diagnosis of CRC by training on a large data set of white-light colonoscopy images and achieve endoscopist-level performance on three independent datasets.

Drought stress limits wheat production around the world. Here, the authors report allelic variation of a WD40 encoding gene TaWD40-4B.1 contributes to wheat drought tolerance by interacting with canonical catalases as a chaperone to promote their activities to avoid ROS over accumulation.

Deep learning methods can provide useful predictions for drug design, but their hyperparameters need to be carefully tweaked to give good performance on a specific problem or dataset. Li et al. present here a method that finds appropriate architectures and hyperparameters for a wide range of drug design tasks and can achieve good performance without human intervention.

Mechanisms and consequences of the acclimation of soil respiration to warming are unclear. Here, the authors combine soil respiration, metagenomics, and functional gene results from a 7-year grassland warming experiment to a microbial-enzyme decomposition model, showing functional gene information to lower uncertainty and improve fit.

Designing molecules in drug design is challenging as it requires optimizing multiple, potentially competing qualities. Wu and colleagues present a prompt-based molecule optimization method that can be trained from single-property data.

The study identifies TaSPL6-D as a negative regulator of TaHKT1;5-D and salinity tolerance in bread wheat. An insertion variation of TaSPL6-D, mainly hidden in landraces, shows the potential for breeding salt-tolerant crops.

Combining generative models and reinforcement learning has become a promising direction for computational drug design, but it is challenging to train an efficient model that produces candidate molecules with high diversity. Jike Wang and colleagues present a method, using knowledge distillation, to condense a conditional transformer model to make it usable in reinforcement learning while still generating diverse molecules that optimize multiple molecular properties.

The lymphotoxin (LT) system has several defined roles in the regulation of immunity as well as vascular and lymphatic biology, however its role in cancer is less understood. Here the authors show that regulatory T cell-derived LTα1β2 promotes LTβ receptor-nonclassical NFκB signaling in tumor cells and lymphatic endothelial cells, accelerating tumor growth and metastasis.

A study using paired ¹⁵N tracers shows atmospheric N deposited in oxidized form is more likely retained by trees, while the reduced form is retained in soil. The authors argue that this is a greater contribution of deposited N to the global forest C sink than previously reported.

Wang et al. propose EasIFA, an efficient enzyme active site annotation algorithm, to advance various fields including drug discovery, disease research, enzyme engineering, and synthetic biology.

Attempts to explain molecular property predictions of neural networks are not always compatible with chemical intuition based on chemical substructures. Here the authors propose the substructure mask explanation method to tackle this challenge.

A high-quality rubber tree genome reveals insights into the evolution of rubber biosynthesis and ethylene stimulation in rubber production. Together with transcriptome data, this study provides valuable data for the research and breeding of rubber trees.

Drug repositioning offers a promising avenue for accelerating drug development, yet existing methods struggle with network diversity, cold start issues, and intrinsic attribute representation. Here, the authors introduce UKEDR, a deep learning framework that integrates knowledge graph embedding and pre-training strategies to overcome the intractable cold start issue, achieving superior performance and interpretability in drug repurposing.

Mobile tet(X3) and tet(X4) genes are identified on conjugative plasmids in Enterobacteriaceae and Acinetobacter isolated from humans, meat for consumption and animals that confer resistance to tetracyclines, including tigecycline, eravacycline and omadacycline.

Polyurethane ionogels can be made with UV light curing but suffer from low mechanical properties compared to conventional polyurethane. Here, robust UV light curable ionogels are made by using prepolymers containing carbon-carbon double bonds as monomers which cross-links to form entangled polymer networks.