AbraLlama: Predicting Abraham Model Solute Descriptors and Modified Solvent Parameters Using Llama

Liquids, Год журнала: 2024, Номер 4(3), С. 518 - 524

Опубликована: Авг. 2, 2024

This study explores the application of fine-tuned large language models for predicting physicochemical properties, specifically focusing on Abraham model solute descriptors (E, S, A, B, V) and modified solvent parameters (e0, s0, a0, b0, v0). By leveraging ChemLLaMA, a specialized version LLaMA cheminformatics tasks, we developed AbraLlama-Solvent AbraLlama-Solute using curated datasets experimentally derived parameters. Our findings demonstrate that predict with high accuracy, comparable to existing methods. The shows varying prediction accuracy across different solvents, influenced by their position within chemical space, while consistently predicts accuracy. Both are available as applications Hugging Face, facilitating easy predictions from SMILES strings. research highlights potential LLMs in chemistry applications, offering practical tools comparison expanding applicability solvation equations broader range organic solvents.

Язык: Английский

2-Methyl tetrahydrofuran: a green organic modifier for eco-friendly, cost-effective, and rapid purification in drug discovery

Analytical Methods, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

2-Methyltetrahydrofuran (MTHF) improved separation efficiency, sample loading, and purification, reducing solvent use by 87% process time 89%. It enhanced loading capacity solubility, offering a sustainable, cost-effective solution.

Язык: Английский

Designing novel waterborne polylactic acid coatings for food packaging applications including polyvinyl alcohol as stabilizer

Food Packaging and Shelf Life, Год журнала: 2025, Номер 49, С. 101493 - 101493

Опубликована: Апрель 8, 2025

Язык: Английский