CompChem Highlights

Important recent papers in computational and theoretical chemistry
A free resource for scientists run by scientists

[CompChem Highlights @CCHighlights]

Lifelong Machine Learning Potentials https://t.co/28XTKGHvbU #compchem

[CompChem Highlights @CCHighlights]

Toy Models of Superposition https://t.co/opkd7ljdie

[CompChem Highlights @CCHighlights]

Variational Pair-Density Functional Theory: Dealing with Strong Correlation at the Protein Scale https://t.co/P2HPdm8czG #compchem

[CompChem Highlights @CCHighlights]

Reproducing Reaction Mechanisms with Machine Learning Models Trained on a Large-Scale Mechanistic Dataset https://t.co/wGUx8GDhg0 #compchem

[CompChem Highlights @CCHighlights]

Using GNN property predictors as molecule generators https://t.co/FgdlkKkouI #compchem

[CompChem Highlights @CCHighlights]

FragGT: Fragment-based evolutionary molecule generation with gene types https://t.co/spRl9QyIhd #compchem

[CompChem Highlights @CCHighlights]

@wangleiphy Except that SELFIES were invented specifically to always give valid molecules. If you filter out molecules, then the raison d'être for SELFIES (which arguably has a syntax that is more difficult to learn) disappears.

[CompChem Highlights @CCHighlights]

@gklambauer If SELFIES is a harder syntax to learn, isn't it all the more important to remove the cases where it goes wrong?

[CompChem Highlights @CCHighlights]

Invalid SMILES are beneficial rather than detrimental to chemical language models https://t.co/P9XDwuhfVH #compchem

[CompChem Highlights @CCHighlights]

An evolutionary algorithm for interpretable molecular representations https://t.co/n2eo6YOdqg #compchem

[CompChem Highlights @CCHighlights]

AiZynth Impact on Medicinal Chemistry Practice at AstraZeneca https://t.co/zAN8kjQioG #compchem

[CompChem Highlights @CCHighlights]

TS-Tools: Rapid and Automated Localization of Transition States Based on a Textual Reaction SMILES Input https://t.co/FYQg7HaVT2 #compchem

[CompChem Highlights @CCHighlights]

Accurate transition state generation with an object-aware equivariant elementary reaction diffusion model https://t.co/JvtN1sGkps #compchem

[CompChem Highlights @CCHighlights]

Growing strings in a chemical reaction space for searching retrosynthesis pathways https://t.co/sStrc0Bn9S #compchem

[CompChem Highlights @CCHighlights]

Few-Shot Learning for Low-Data Drug Discovery
https://t.co/OQoTOZLcZx #compchem

[CompChem Highlights @CCHighlights]

Ranking Pareto optimal solutions based on projection free energy https://t.co/ackF6LNcUs #compchem

[CompChem Highlights @CCHighlights]

Accelerated dinuclear palladium catalyst identification through unsupervised machine learning https://t.co/27aPEt81WP #compchem

[CompChem Highlights @CCHighlights]

Real-World Molecular Out-Of-Distribution: Specification and Investigation https://t.co/W3FMOnyih8 #compchem

[CompChem Highlights @CCHighlights]

Evolutionary Multiobjective Optimization of Multiligand Metal Complexes in Diverse and Vast Chemical Spaces https://t.co/r35wvfz5V3 #compchem

[CompChem Highlights @CCHighlights]

Seeing is Believing: Brain-Inspired Modular Training for Mechanistic Interpretability https://t.co/Kdpay5W5sE #compchem