A5015848532- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Explainable Artificial Intelligence (XAI)
- Essential Oils and Antimicrobial Activity
- Machine Learning in Healthcare
- Synthesis and biological activity
- Advanced Graph Neural Networks
- Natural product bioactivities and synthesis
- Domain Adaptation and Few-Shot Learning
- Topic Modeling
- Click Chemistry and Applications
- Chemical Synthesis and Analysis
- Phytochemistry and Biological Activities
- SARS-CoV-2 and COVID-19 Research
- Machine Learning and ELM
- Chemistry and Chemical Engineering
- Traffic Prediction and Management Techniques
- Malaria Research and Control
- Advanced Memory and Neural Computing
- Plant Pathogens and Fungal Diseases
- Antibiotic Resistance in Bacteria
Sapienza University of Rome
2021-2025
Centre National de la Recherche Scientifique
2025
Laboratoire d'Informatique en Images et Systèmes d'Information
2025
Institut National des Sciences Appliquées de Lyon
2025
Essential oils (EOs) exhibit a broad spectrum of biological activities; however, their clinical application is hindered by challenges, such as variability in chemical composition and chemical/physical instability. A critical limitation the lack consistency across EO samples, which impedes standardization. Despite this, evidence suggests that EOs with differing profiles often display similar (micro)biological activities, raising possibility standardizing based on effects rather than...
Graph neural networks have proved to be a key tool for dealing with many problems and domains such as chemistry, natural language processing social networks. While the structure of layers is simple, it difficult identify patterns learned by graph network. Several works propose post-hoc methods explain predictions, but few them try generate interpretable models. Conversely, topic models highly investigated in image recognition. Given similarity between domains, we analyze adaptability...
Abstract The main protease (M pro ) of SARS-Cov-2 is the essential enzyme for maturation functional proteins implicated in viral replication and transcription. peculiarity its specific cleavage site joint with high degree conservation among all coronaviruses promote it as an attractive target to develop broad-spectrum inhibitors, selectivity tolerable safety profile. Herein reported a combination three-dimensional quantitative structure–activity relationships (3-D QSAR) comparative molecular...
Abstract Molecular property prediction is a fundamental task in the field of drug discovery. Several works use graph neural networks to leverage molecular representations. Although they have been successfully applied variety applications, their decision process not transparent. In this work, we adapt concept whitening networks. This approach an explainability method used build inherently interpretable model, which allows identifying concepts and consequently structural parts molecules that...
Scientific investigation on essential oils composition and the related biological profile are continuously growing. Nevertheless, only a few studies have been performed relationships between chemical data. Herein, of 61 assayed is reported focusing their inhibition activity against
Field-Based Three-Dimensiona Quantitative Strucutere-Activity Relationships (FB 3D QSAR) comprise computational approaches used in drug design and molecular modeling to analyze the relationship between three-dimensional structure of a list molecules (described by interaction fields) their associated biological activities (BAs). It aims understand how different structural features contribute enhancing or lowering potency. The process FB QSAR involves several steps. First, dataset structurally...
Two of the most impressive features biological neural networks are their high energy efficiency and ability to continuously adapt varying inputs. On contrary, amount power required train top-performing deep learning models rises as they become more complex. This is main reason for increasing research interest in spiking networks, which mimic functioning human brain achieving similar performances artificial but with much lower costs. However, even this type network not provided incrementally...
Pharmacological properties of essential oils Pinus species are mainly associated with antimicrobial, antioxidant, anticancer, anti-aging, and anti-inflammatory potencies. However, only limited scientific information has been gathered regarding the genotoxic antigenotoxic activities oils. Therefore, aim present study was to investigate in vitro DNA protective three commercial species: P. mugo, sibirica, silvestre, against oxidative damage induced by hydroxyl peroxyl radicals. The tested...
It is well known that Drug Design often a costly process both in terms of time and economic effort. While good Quantitative Structure-Activity Relationship models (QSAR) can help predicting molecular properties without the need to synthesize them, it still required come up with new molecules be tested. This mostly done lack tools determine which modifications are more promising or aspects molecule influential for final activity/property. Here we present an automatic involves Graph...
Since the introduction of artificial intelligence in medicinal chemistry, necessity has emerged to analyse how molecular property variation is modulated by either single atoms or chemical groups. In this paper, we propose train graph-to-graph neural network using semi-supervised learning for attributing structure-property relationships. As initial case studies apply method solubility and acidity while checking its consistency comparison with known experimental data. final goal, our approach...