Rewrote the existing C code for a simple molecular dynamics simulation in Rust.
What a brilliant language. So clean and clear.
Fun experience. Took me 3 days. (No LLM help haha).
Unlocking Performance: Benchmarking Molecular Dynamics Simulations with Advanced Compiler Optimizations
In the realm of scientific computing, the efficiency of molecular dynamics simulations can make or break research outcomes. This article delves into the intricacies of benchmarking a calcium aluminosi...
Running 16 simulations in parallel is a very efficient way of making simulations.
It sucks however, that they all crash if one of them crashes.
Young Scientist Award:
Petrick Heighway
for his pivotal role in the DiPOLE-100x Community Proposals measuring #XRay #diffraction at extreme pressures and temperatures at the HED-HiBEF instrument.
Combines experimental data with #MolecularDynamics simulations to provide critical insight into the nature of release pathways of shock compressed materials, kinematics of plastisticy, …
I am shamefully copying this post from Linkedin but if you happen to know any candidate, do not hesitate to reach:
It's on. To my #ComputationalChemistry network, I am calling on you. 
The PM2E team in the CIMAP laboratory (www.cimap.ensicaen.fr) is looking for a #PhD hashtag#candidate. The goal of the project is to understand key parameters in #semiconductor defects stability and physics (both vacancies and color centers, see for example: https://doi.org/10.1016/j.nimb.2022.12.025) through the use of
#DFT and #MolecularModeling. The end goals of the project is concerned with application for #QuantumComputing through the making of stable #qubit. Collaboration and communication with experimentalists will be part of the project.
If you're interested in #MolecularDynamics, #MonteCarlo simulations, or DFT and quantum calculations, we are waiting for you. We are using codes such as #Lammps and #QuantumEspresso. We are looking for a candidate finishing or already having a master degree (or equivalent diploma) with training in #physics/ #MaterialsScience / #TheoreticalChemistry. Any experience in programming using #Python, #C and the #Linux environment would be considered a plus.
The position is located in Alençon (Orne) between Caen and Le Mans. Founding is assured for relevant candidates.
Link to the offer (in french, you can ask for an english version): https://www.abg.asso.fr/en/candidatOffres/show/id_offre/127985/job/simulation-atomistique-de-defauts-crees-par-irradiation-dans-les-semiconducteurs
Trying to get back in the #moleculardynamics bandwagon. Found a #NMR structure for a relatively small neurotoxin protein bound to a small peptide that neutralizes it. No one seems to have done any MD simulations yet. Started some sims. Maybe I’ll observe something interesting worth publishing.
Is anyone doing structure prediction with #AlphaFold3 figured out an easy way to do the relaxation step that Colabfold did? I'm looking into OpenMM and Amber, and making some progress, but it seems like it will be very difficult to make it work when phosphoresidues, ions, or anything else besides standard amino acid residues are included.
The D. E. Shaw Research DNA/RNA #moleculardynamics force-fields look great.
These non-canonical DNA structures are beautiful. Observing B-DNA <-> Z-DNA inter-conversion during simulation is so cool.
G-quadruplexes are such neat compact bundles.
The deficiencies are in nucleic acid-protein interactions -- getting values from different force-fields to play nice and be consistent is hard.
• Paper: https://pubs.acs.org/doi/10.1021/acs.jpcb.1c10971
• Force-field code: https://github.com/DEShawResearch/viparr-ffpublic
https://pubs.acs.org/doi/10.1021/acs.jctc.4c00773
A great summary of the current improvement in #RNA force-fields in #moleculardynamics, with a table of applied case studies and a section on incorporating experimental data
In-Situ Techniques on GPU-Accelerated Data-Intensive Applications
RBMD: A molecular dynamics package enabling to simulate 10 million all-atom particles in a single graphics processing unit
Para los que estuvieron en mi charla en la FLISoL sobre clústers de cálculo numérico, hoy largamos 729 simulaciones de 3000 partículas al clúster de una, más o menos esperamos que el mismo complete 10 corridas por día, por lo que en unos 73 días tendríamos los resultados #hpc #cluster #ciencia #md #dinamicamolecular #moleculardynamics #science #fisica #physics
Gaining Cross-Platform Parallelism for HAL’s Molecular Dynamics Package using SYCL
J'ai fait une petite simulations rigolote pour vérifier numériquement un vieux truc que j'ai trouvé dans un article de Boltzmann. Si on met une particule dans un plan avec deux ressorts qui ont des constantes de forces k1 et k2, la trajectoire est périodique si k1/k2 est un nombre rationnel: https://youtu.be/keGGbb2uiB8
Évidemment si c'est pas un nombre rationnel (genre pi), c'est pas périodique: https://youtu.be/Tty0jFWKkv8
Voilà, c'est tout pour moi.
GROMACS on AMD GPU-Based HPC Platforms: Using SYCL for Performance and Portability
#Physics #MolecularDynamics #HPC #Performance #SYCL #Package
Molecular dynamics and multiscale modelling for sustainability
DAIMON Team - CNR-ISMN
Full time researcher in #moleculardynamics and #multiscalemodelling for sustainable materials @CNR-ISMN Bologna, Italy #Gromacs #LAMMPS
See the full job description on jobRxiv: https://jobrxiv.org/job/daimon-team-cnr-ismn-27778-molecular-dynamics-and-multiscale-modelling-for-su...
https://jobrxiv.org/job/daimon-team-cnr-ismn-27778-molecular-dynamics-and-multiscale-modelling-for-sustainability/?feed_id=69895
A significant challenge of biomolecular simulations are the computational limitations. A new method aims to overcome these by employing #Bayesian neural network learning models.
More details in the #preprint evaluated on @prereview
Happy to see another command I worked on merged in the LAMMPS code.
Langevin barostat is now available to use, thanks to the dev team which is of very wise help.
Waiting for users feedback, if any. 
