4 598
订阅者
+624 小时
+147 天
+6730 天
帖子存档
Two German doctoral students were studying an unrelated reaction when their flasks filled with strange orange and yellow solids, and they had accidentally made molecules chemists had chased since the 1960s - https://share.google/PFOjB1bSndyoNTAdM
14 11月, 06:05
How Avogadro's Number Was Discovered
https://youtube.com/watch?v=QLnQFthLYi4&is=gekZkZ-RFxSpfw7J
10 11月, 04:25
One force field for proteins, DNA, lipids and carbohydrates?
Happy to share SO3LR Tauri, a pretrained machine-learned force field that couples a fast neural network with explicit repulsion, electrostatics and dispersion. It takes a step toward a unified model that describes all major classes of (bio)molecules at quantum-mechanical accuracy.
We benchmarked the model across a broad range of systems to assess its capabilities and limitations. We simulated the classes mentioned above, as well as a platinum-DNA complex, organic and enzyme-related reaction pathways, pKas, water and organic liquids, ions, protein-ligand complexes and MOFs, all without system-specific fitting. Released in three sizes, the models scale to 27-84M atoms and reach 0.7-3.2 μs/atom on a single GPU.
We also discuss many challenges that we encountered: liquid transport properties, broader coverage of biomolecular conformations in QM training data, and longer sampling of slow biomolecular processes (which I hope can be circumvented with a genAI+MLFF combo). Thrilled to share it!
preprint: https://arxiv.org/abs/2611.06237v1
code: https://github.com/general-molecular-simulations/so3lr
docs: https://so3lr.github.io/
data: https://zenodo.org/records/22920714
md example: so3lr nvt --input geom.xyz
06 11月, 07:35
🧪 Avogadro 2.1.0 Released!
Avogadro, the free and open-source molecular editor and visualization software, has received a major update, bringing new features for computational chemistry, molecular modeling, and education.
🚀 What's New in Version 2.1.0?
🔹 New Z-Matrix Editor: Edit internal coordinates, apply geometric constraints, and reorganize atoms.
🔹 Improved Geometry Optimization: Implementation of the ABC-FIRE algorithm and a new hybrid optimization mode.
🔹 Up to 25× Faster Performance: Significant improvements in Gaussian basis function calculations, accelerating molecular orbital and surface generation. Faster loading of large ORCA output files.
🔹 New Molecular Modeling Tools: Polymer building, direct editing of bond lengths and angles, and improved support for CIF and PDB files.
🔹 Enhanced Visualization: Improvements in vibrational modes, molecular conformations, trajectories, and molecular properties.
🔹 Python and AI Integration: Expanded JSON-RPC interface, Jupyter automation, and the new
avogadro-mcp project for integration with AI agents.
🔹 Better Stability and Extensibility: Numerous bug fixes and an expanded ecosystem featuring 30 plugins.
💻 Available for Windows, Linux, and macOS.
📥 Download:
https://avogadro.cc/install/
📋 Full Release Notes:
https://avogadro.cc/install/versions/v210.html
🔗 GitHub Release:
https://github.com/OpenChemistry/avogadroapp/releases/tag/2.1.023 10月, 14:04
Digichem 8.1.1 – Computational Chemistry Software
Digichem is a computational chemistry platform designed to simplify the setup, execution, management, and analysis of quantum chemistry calculations. It offers an interactive interface, automated result processing, calculation chaining, and molecular visualizations suitable for publications.
The newly released version 8.1.1 improves xTB support for very large molecular systems, including memory management and out-of-core calculations, and adds new double-hybrid DFT functional variants (SCS/SOS-ωPBEPP86).
The software is available for a free trial.
Website: https://digi-chem.co.uk/
Release notes: https://digi-chem.co.uk/updates/digichem-release-8.1.1
21 10月, 05:35
ORCA 7 is coming!
To celebrate the release, we're hosting an online ORCA 7 (FACCTs & Max-Planck-Institut für Kohlenforschung) Release Event on
December 8 from 4-8 PM CET via Zoom (Link will be posted later).
We'll introduce the new release, highlight key developments and new capabilities, and give you a closer look at what's coming to the ORCA ecosystem.
It's also a chance to hear directly from the team behind ORCA, ask questions, and discuss where we're heading next.
Whether you've been part of the ORCA community for years or are just getting started, we'd be very happy to have you with us.
Save the date and stay tuned for more details!
#ORCA7 #ORCA #FACCTs #MPIKOFO #ComputationalChemistry #QuantumChemistry #TheoreticalChemistry #CompChem #DFT #ScientificSoftware
20 10月, 11:12
🚀 Big news for the computational chemistry and bioinformatics community!
Our team has been working on modernizing MGLTools 1.5.6 to make it compatible with modern computing environments and easier to use for molecular docking and structural biology research.
MGLTools has been an important part of molecular docking workflows for years. However, its legacy Python 2 codebase and older graphics dependencies have made it increasingly challenging to run on modern systems.
With our MGLTools-Py3 project, we have brought MGLTools 1.5.6 to a modern Python environment.
The Windows build is now available, with support for:
- Modern AMD and Intel PCs with native 64-bit support.
- Qualcomm Snapdragon-based Windows 11 PCs through Prism emulation.
- Portable execution: unzip the package and launch
adt.exe, without requiring a separate Python installation.
Key technical updates include:
- Migration to Python 3.13 and NumPy 2.x.
- Modernization of 15 native C/C++ SWIG extensions.
- Replacement of the proprietary MSMS component with an open-source solvent-excluded surface engine.
- Validation of key AutoDock preparation scripts against reference outputs.
- Improvements to OpenGL rendering for molecular surfaces, CPK representations, and secondary structure visualization.
Linux and macOS builds, including Apple Silicon support, are planned.
🔗 GitHub repository: https://lnkd.in/dSrpQsx9
We invite researchers, educators, and developers working in molecular docking, virtual screening, and structural biology to test the project and share their feedback. Community testing will help us improve compatibility and reliability further.
https://www.linkedin.com/posts/angamba_moleculardocking-computationalchemistry-bioinformatics-activity-7515830745974779904-T_mu?utm_source=social_share_send&utm_medium=member_desktop_web&rcm=ACoAADByb54BZAu0zfSJLooSdNdx0bXFCOsvoA0
#MolecularDocking #ComputationalChemistry #Bioinformatics #DrugDiscovery14 10月, 22:12
🚀 Big news for the computational chemistry and bioinformatics community!
Our team has been working on modernizing MGLTools 1.5.6 to make it compatible with modern computing environments and easier to use for molecular docking and structural biology research.
MGLTools has been an important part of molecular docking workflows for years. However, its legacy Python 2 codebase and older graphics dependencies have made it increasingly challenging to run on modern systems.
With our MGLTools-Py3 project, we have brought MGLTools 1.5.6 to a modern Python environment.
The Windows build is now available, with support for:
- Modern AMD and Intel PCs with native 64-bit support.
- Qualcomm Snapdragon-based Windows 11 PCs through Prism emulation.
- Portable execution: unzip the package and launch
adt.exe, without requiring a separate Python installation.
Key technical updates include:
- Migration to Python 3.13 and NumPy 2.x.
- Modernization of 15 native C/C++ SWIG extensions.
- Replacement of the proprietary MSMS component with an open-source solvent-excluded surface engine.
- Validation of key AutoDock preparation scripts against reference outputs.
- Improvements to OpenGL rendering for molecular surfaces, CPK representations, and secondary structure visualization.
Linux and macOS builds, including Apple Silicon support, are planned.
🔗 GitHub repository: https://lnkd.in/dSrpQsx9
We invite researchers, educators, and developers working in molecular docking, virtual screening, and structural biology to test the project and share their feedback. Community testing will help us improve compatibility and reliability further.
hashtagMolecularDocking hashtagComputationalChemistry hashtagBioinformatics hashtagDrugDiscovery hashtag14 10月, 19:01
14 10月, ~19:02
The amazing story behind B3LYP
https://www.henriquecastro.info/2026/10/12/the-strange-pragmatic-birth-of-b3lyp/
14 10月, 04:24
WATOC 2028 is now on social media!
We are pleased to announce that the official social media channels for WATOC 2028 are now active.
Follow us to stay informed about congress news, announcements, important dates, and updates as we prepare for the 14th Triennial Congress of the World Association of Theoretical and Computational Chemists.
Connect with us on our official social media channels:
Instagram: @watoc2028
𝕏: @Watoc2028
LinkedIn: WATOC 2028
Pre-register for WATOC 2028:
https://lnkd.in/gHh_6GAG
Official website:
https://watoc2028.org/
Follow us on all our platforms and be part of the WATOC 2028 community!
#WATOC2028 #TheoreticalChemistry #ComputationalChemistry #Chemistry #ScientificCommunity #Mérida #Mexico
12 10月, 02:05
Chemprop predicts molecular properties from chemical structure using message passing neural networks, with a rewritten v2 release, docs, and tutorial notebooks for research use.
Explore it here:
https://osp.fyi/chemprop
11 10月, 02:20
