Explore the vast range of titles available.

Generative models are becoming a tool of choice for exploring the molecular space. These models learn on a large training dataset and produce novel molecular structures with similar properties. Generated structures can be utilized for virtual screening or training semi-supervized predictive models in the downstream tasks. While there are plenty of generative models, it is unclear how to compare and rank them. In this work, we introduce a benchmarking platform called Molecular Sets (MOSES) to standardize training and comparison of molecular generative models. MOSES provides training and testing datasets, and a set of metrics to evaluate the quality and diversity of generated structures. We have implemented and compared several molecular generation models and suggest to use our results as reference points for further advancements in generative chemistry research. The platform and source code are available at https://github.com/molecularsets/moses .

Uperin 3.5 is a remarkable natural peptide obtained from the skin of toadlets comprised of 17 amino acids which exhibits both antimicrobial and amyloidogenic properties. Molecular dynamics simulations were performed to study the β-aggregation process of uperin 3.5 as well as two of its mutants, in which the positively charged residues Arg7 and Lys8 have been replaced by alanine. All three peptides rapidly underwent spontaneous aggregation and conformational transition from random coils to beta-rich structures. The simulations reveal that the initial and essential step of the aggregation process involves peptide dimerization and the formation of small beta-sheets. A decrease in positive charge and an increase in the number of hydrophobic residues in the mutant peptides lead to an increase in the rate of their aggregation.

The interaction of antimicrobial and amyloid peptides with cell membranes is a critical step in their activities. Peptides of the uperin family obtained from the skin secretion of Australian amphibians demonstrate antimicrobial and amyloidogenic properties. All-atomic molecular dynamics and an umbrella sampling approach were used to study the interaction of uperins with model bacterial membrane. Two stable configurations of peptides were found. In the bound state, the peptides in helical form were located right under the head group region in parallel orientation with respect to the bilayer surface. Stable transmembrane configuration was observed for wild-type uperin and its alanine mutant in both alpha-helical and extended unstructured forms. The potential of mean force characterized the process of peptide binding from water to the lipid bilayer and its insertion into the membrane, and revealed that the transition of uperins from the bound state to the transmembrane position was accompanied by the rotation of peptides and passes through the energy barrier of 4–5 kcal/mol. Uperins have a weak effect on membrane properties.

We have studied the molecular properties of water in physically and chemically cross-linked gelatin hydrogels by FTIR-spectroscopy, NMR relaxation, and diffusivity and broadband dielectric spectroscopy, which are sensitive to dynamical properties of water, being a structural marker of polymer network. All experiments demonstrated definite reinforcement of the hydrogel net structure and an increase in the amount of hydrate water. FTIR experiments have shown that the chemical cross-linking of gelatin molecules initiates an increase in the collagen-like triple helices “strength”, as a result of infused restriction on protein molecular mobility. The “strengthening” of protein chains hinders the mobility of protein fragments, introducing complex modifications into the structural properties of water which are remained practically unchanged up to up to 30–40 °C.

Natural-language-to-visualization (NL2VIS) systems based on large language models (LLMs) have substantially improved the accessibility of data visualization. However, their further adoption is hindered by two coupled challenges: (i) the absence of standardized evaluation metrics makes it difficult to assess progress in the field and compare different approaches; and (ii) natural language descriptions are inherently underspecified, so multiple visualizations may be valid for the same query. To address these issues, we introduce VegaChat, a framework for generating, validating, and assessing declarative visualizations from natural language. We propose two complementary metrics: Spec Score, a deterministic metric that measures specification-level similarity without invoking an LLM, and Vision Score, a library-agnostic, image-based metric that leverages a multimodal LLM to assess chart similarity and prompt compliance. We evaluate VegaChat on the NLV Corpus and on the annotated subset of ChartLLM. VegaChat achieves near-zero rates of invalid or empty visualizations, while Spec Score and Vision Score exhibit strong correlation with human judgments (Pearson 0.65 and 0.71, respectively), indicating that the proposed metrics support consistent, cross-library comparison. The code and evaluation artifacts are available at https://zenodo.org/records/17062309.

Generative models are becoming a tool of choice for exploring the molecular space. These models learn on a large training dataset and produce novel molecular structures with similar properties. Generated structures can be utilized for virtual screening or training semi-supervised predictive models in the downstream tasks. While there are plenty of generative models, it is unclear how to compare and rank them. In this work, we introduce a benchmarking platform called Molecular Sets (MOSES) to standardize training and comparison of molecular generative models. MOSES provides a training and testing datasets, and a set of metrics to evaluate the quality and diversity of generated structures. We have implemented and compared several molecular generation models and suggest to use our results as reference points for further advancements in generative chemistry research. The platform and source code are available at https://github.com/molecularsets/moses.