A5032123673- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Protein Structure and Dynamics
- Chemical Synthesis and Analysis
- Neuroscience and Neuropharmacology Research
- Cannabis and Cannabinoid Research
- Tryptophan and brain disorders
- Receptor Mechanisms and Signaling
- Various Chemistry Research Topics
- Chemistry and Chemical Engineering
- Bioinformatics and Genomic Networks
- Innovative Microfluidic and Catalytic Techniques Innovation
- Click Chemistry and Applications
- Neuroinflammation and Neurodegeneration Mechanisms
- X-ray Diffraction in Crystallography
- Scientific Computing and Data Management
- Neurotransmitter Receptor Influence on Behavior
- Crystallization and Solubility Studies
- Free Radicals and Antioxidants
- Pancreatic function and diabetes
- Pharmacological Receptor Mechanisms and Effects
- Immunotherapy and Immune Responses
- Catalytic C–H Functionalization Methods
- Cancer, Stress, Anesthesia, and Immune Response
- Luminescence and Fluorescent Materials
Roche (Switzerland)
2019-2025
ETH Zurich
2020-2025
University of Basel
2019-2020
Generative chemical language models (CLMs) can be used for de novo molecular structure generation by learning from a textual representation of molecules. Here, we show that hybrid CLMs additionally leverage the bioactivity information available training compounds. To computationally design ligands phosphoinositide 3-kinase gamma (PI3Kγ), collection virtual molecules was created with generative CLM. This compound library refined using CLM-based classifier prediction. second CLM pretrained...
Abstract Machine learning approaches in drug discovery, as well other areas of the chemical sciences, benefit from curated datasets physical molecular properties. However, there currently is a lack data collections featuring large bioactive molecules alongside first-principle quantum information. The open-access QMugs (Quantum-Mechanical Properties Drug-like Molecules) dataset fills this void. collection comprises mechanical properties more than 665 k biologically and pharmacologically...
Abstract De novo drug design aims to generate molecules from scratch that possess specific chemical and pharmacological properties. We present a computational approach utilizing interactome-based deep learning for ligand- structure-based generation of drug-like molecules. This method capitalizes on the unique strengths both graph neural networks language models, offering an alternative need application-specific reinforcement, transfer, or few-shot learning. It enables “zero-shot"...
Generative deep learning and miniaturized bench-top synthesis were combined to automate the design of novel LXR agonists.
Many molecular design tasks benefit from fast and accurate calculations of quantum-mechanical (QM) properties. However, the computational cost QM methods applied to drug-like molecules currently renders large-scale applications quantum chemistry challenging. Aiming mitigate this problem, we developed DelFTa, an open-source toolbox for prediction electronic properties at density functional (DFT) level theory, using Δ-machine-learning. Δ-Learning corrects error (Δ) a but inaccurate property...
Abstract Late-stage functionalization is an economical approach to optimize the properties of drug candidates. However, chemical complexity molecules often makes late-stage diversification challenging. To address this problem, a platform based on geometric deep learning and high-throughput reaction screening was developed. Considering borylation as critical step in functionalization, computational model predicted yields for diverse conditions with mean absolute error margin 4–5%, while...
A deep learning approach centered on electron density is suggested for predicting the binding affility between proteins and ligands. The thoroughly assessed using various pertinent benchmarks.
Detection and visualization of the cannabinoid receptor type 2 by a cell-permeable high affinity fluorescent probe platform enables tracing trafficking in live cells zebrafish.
We introduce a design principle to stabilize helically chiral structures from an achiral tetrasubstituted [2.2]paracyclophane by integrating it into macrocycle. The introduces three-dimensional perturbation nearly planar macrocyclic oligothiophene. resulting helical structure is stabilized two bulky substituents installed on the unit. increased enantiomerization barrier enabled separation of both enantiomers. synthesis target macrocycle 1 involves sequence halogenation and cross-coupling...
Despite the broad implications of cannabinoid type 2 receptor (CB2) in neuroinflammatory processes, a suitable CB2-targeted probe is currently lacking clinical routine. In this work, we synthesized 15 fluorinated pyridine derivatives and tested their binding affinities toward CB2 CB1. With sub-nanomolar affinity (
Pharmacological modulation of cannabinoid type 2 receptor (CB2R) holds promise for the treatment numerous conditions, including inflammatory diseases, autoimmune disorders, pain, and cancer. Despite significance this receptor, researchers lack reliable tools to address questions concerning expression complex mechanism CB2R signaling, especially in cell-type tissue-dependent contexts. Herein, we report first time a versatile ligand platform modular design collection highly specific...
The rapid and economical synthesis of novel bioactive compounds remains a significant hurdle in drug discovery efforts. This study demonstrates an integrated medicinal chemistry workflow that effectively diversifies hit lead structures, enabling efficient acceleration the critical hit-to-lead optimization phase. Employing high-throughput experimentation (HTE), we generated comprehensive data set encompassing 13,490 Minisci-type C-H alkylation reactions. served as foundation for training deep...
Abstract Utilizing the growing wealth of chemical reaction data can boost synthesis planning and increase success rates. Yet, effectiveness machine learning tools for retrosynthesis forward prediction relies on accessible, well‐curated presented in a structured format. Although some public licensed databases exist, they often lack essential information about conditions. To address this issue promote principles findable, interoperable, reusable (FAIR) reporting sharing, we introduce Simple...
ATP synthase dysregulation has been implicated in many diseases, including cancer and neurodegenerative diseases. Whilst synthase‐targeting compounds have reported, most are large or polar lack appropriate properties for a CNS drug. We designed, synthesised, evaluated novel series of targeting compounds, resulting 1,3,4‐oxadiazin‐5‐one scaffold with improved physiochemical properties. In vitro evaluation our library led to the discovery CJ1 34 as partial inhibitor determined IC50 394 nM...
The rapid and economical synthesis of novel bioactive compounds remains a significant hurdle in drug discovery efforts. This study demonstrates an integrated medicinal chemistry workflow that effectively diversifies hit lead structures, enabling efficient acceleration the critical hit-to-lead optimization phase. Employing high-throughput experimentation (HTE), we generated comprehensive data set encompassing 13,490 Minisci-type C-H alkylation reactions. served as foundation for training deep...
Leveraging the increasing volume of chemical reaction data can enhance synthesis planning and improve suc- cess rates. However, machine learning applications for retrosynthesis forward prediction tools depend on having readily available, high-quality in a structured format. While some public licensed databases are they frequently lack essential information about condi- tions. To address this issue promote principles findable, accessible, interoperable, reusable (FAIR) reporting sharing, we...
Protein-ligand interactions (PLIs) determine the efficacy and safety profiles of small molecule drugs. Existing methods rely on either structural information or resource-intensive computations to predict PLI, casting doubt whether it is possible perform structure-free PLI predictions at low computational cost. Here we show that a light-weight graph neural network (GNN), trained with quantitative PLIs number proteins ligands, able strength unseen PLIs. The model has no direct access about...
The cannabinoid type 2 (CB2) receptor has emerged as a valuable target for therapy and imaging of immune-mediated pathologies. With the aim to find suitable radiofluorinated analogue previously reported CB2 positron emission tomography (PET) radioligand [11C]RSR-056, 38 fluorinated derivatives were synthesized tested by in vitro binding assays. Ki (hCB2) 6 nM selectivity factor nearly 700 over 1 receptors, compound 3 exhibited optimal properties was selected evaluation PET radioligand....
Automation of the molecular design-make-test-analyze cycle speeds up identification hit and lead compounds for drug discovery. Using deep learning computational design a customized microfluidics platform on-chip compound synthesis, liver X receptor (LXR) agonists were generated from scratch. The pipeline was tuned to explore chemical space defined by known LXRα agonists, suggest structural analogs ligands novel cores. To further lead-like molecules ensure compatibility with automated this...
Enhancing the properties of advanced drug candidates is aided by direct incorporation specific chemical groups, avoiding need to construct entire compound from ground up. Nevertheless, their intricacy often poses challenges in predicting reactivity for C-H activation reactions and planning synthesis. We adopted a reaction screening approach that combines high-throughput experimentation (HTE) at nanomolar scale with computational graph neural networks (GNNs). This aims identify suitable...