A5060276392- Computational Drug Discovery Methods
- Metabolomics and Mass Spectrometry Studies
- Genetics, Bioinformatics, and Biomedical Research
- Biomedical Text Mining and Ontologies
- Analytical Chemistry and Chromatography
- Machine Learning in Materials Science
- Protein Structure and Dynamics
- History and advancements in chemistry
- Monoclonal and Polyclonal Antibodies Research
- Water Quality and Pollution Assessment
- Microbial Natural Products and Biosynthesis
- Spectroscopy and Chemometric Analyses
- Machine Learning in Healthcare
- Explainable Artificial Intelligence (XAI)
- Risk and Safety Analysis
- Quality and Safety in Healthcare
- Chemistry and Chemical Engineering
- Cholinesterase and Neurodegenerative Diseases
- Complex Network Analysis Techniques
- Pharmacogenetics and Drug Metabolism
- Water Quality Monitoring and Analysis
- Advanced Causal Inference Techniques
- Space Satellite Systems and Control
- Manufacturing Process and Optimization
- Environmental Toxicology and Ecotoxicology
Dompé (Italy)
2021-2025
Farmaceutici Damor (Italy)
2023-2024
SAS Institute (United States)
2022
Solvay (France)
2019-2021
Université de Strasbourg
2019-2021
Chimie de la Matière Complexe
2019
Centre National de la Recherche Scientifique
2019
Chimie Moléculaire, Macromoléculaire, Matériaux
2019
Institut de Chimie
2019
University of Insubria
2017
la diffusion de documents scientifiques niveau recherche, publiés ou non, émanant des établissements d'enseignement et recherche français étrangers, laboratoires publics privés.
The objective of this study is to evaluate the accuracy and robustness three exposure-modelling tools [STOFFENMANAGER® v.6, European Centre for Ecotoxicology Toxicology Chemical Target Risk Assessment v.3.1 (ECETOC TRA v.3.1), Advanced REACH Tool (ART v.1.5)], by comparing available measured data exposure organic solvents pesticides in occupational scenarios (ESs).Model was evaluated predicted values, expressed as an underestimation or overestimation factor (PRED/EXP), regression analysis....
We report predictive models of acute oral systemic toxicity representing a follow-up our previous work in the framework NICEATM project. It includes update original through addition new data and an external validation using dataset relevant for chemical industry context. A regression model LD50 multi-class classification classes according to Global Harmonized System categories were prepared. ISIDA descriptors used encode molecular structures. Machine learning algorithms included support...
We report new consensus models estimating acute toxicity for algae, Daphnia and fish endpoints. assembled a large collection of 3680 public unique compounds annotated by, at least, one experimental value the given endpoint. Support Vector Machine were internally externally validated following OECD principles. Reasonable predictive performances achieved (RMSEext = 0.56–0.78) which are in line with those state-of-the-art models. The known structural alerts compared analysis atomic...
The bioconcentration factor (BCF), a key parameter required by the REACH regulation, estimates tendency for xenobiotic to concentrate inside living organisms. In silico methods can be valid alternatives costly data measurements. However, in industrial context, these theoretical approaches may fail predict BCF with reasonable accuracy. We analyzed whether models built on public only have adequate performances when challenged compounds. A new set of 1129 compounds has been collected merging...
The scaffold representation is widely employed to classify bioactive compounds on the basis of common core structures or correlate compound classes with specific biological activities. In this paper, we present a novel approach called "Molecular Anatomy" as flexible and unbiased molecular scaffold-based metrics cluster large set compounds. We introduce nine representations at different abstraction levels, combined fragmentation rules, define multi-dimensional network hierarchically...
Abstract Off-target drug interactions are a major reason for candidate failure in the discovery process. Anticipating potential drug’s adverse effects early stages is necessary to minimize health risks patients, animal testing, and economical costs. With constantly increasing size of virtual screening libraries, AI-driven methods can be exploited as first-tier tools provide liability estimation candidates. In this work we present ProfhEX, an suite 46 OECD-compliant machine learning models...
The European Registration, Evaluation, Authorization and Restriction of Chemical Substances Regulation, requires marketed chemicals to be evaluated for Ready Biodegradability (RB), considering in silico prediction as valid alternative experimental testing. However, currently available models may not relevant predict compounds industrial interest, due accuracy applicability domain restriction issues. In this work, we present a new extended RB dataset (2830 compounds), issued by the merging...
Drug-induced cardiotoxicity represents one of the most critical safety concerns in early stages drug development. The blockade human ether-à-go-go-related potassium channel (hERG) is frequent cause cardiotoxicity, as it associated to long QT syndrome which can lead fatal arrhythmias. Therefore, assessing hERG liability new drugs candidates crucial avoid undesired cardiotoxic effects. In this scenario, computational approaches have emerged useful tools for development predictive models able...
The prediction of drug metabolism is attracting great interest for the possibility discarding molecules with unfavorable ADME/Tox profile at early stage discovery process. In this context, artificial intelligence methods can generate highly performing predictive models if they are trained by accurate metabolic data. MetaQSAR-based datasets were collected to predict sites most reactions. based on a set structural, physicochemical, and stereo-electronic descriptors generated random forest...
The evaluation of persistency chemicals in environmental media (water, soil, sediment) is included European Regulations, the context Persistence, Bioaccumulation and Toxicity (PBT) assessment. In silico predictions are valuable alternatives for compounds screening prioritization. However, already existing prediction tools have limitations: narrow applicability domains due to their relatively small training sets, lack medium-specific models. A dataset 1579 unique has been collected, merging...
In the framework of REACH (Registration Evaluation Authorization and restriction Chemicals) regulation, industries have generated reported a huge amount (eco)toxicological data on substance produced or imported in Europe. The registration procedure initiated creation large database well defined properties. Here, distribution chemical space was analyzed with help Generative Topographic Mapping (GTM) approach. GTM generates 2-dimensional maps which each compound is represented as point. 3rd...
Nowadays there is a big spotlight cast on the development of techniques explainable machine learning. Here we introduce new computational paradigm based Group Equivariant Non-Expansive Operators, that can be regarded as product rising mathematical theory information-processing observers. This approach, adjusted to different situations, may have many advantages over other common tools, like Neural Networks, such as: knowledge injection and information engineering, selection relevant features,...
This paper is devoted to the analysis of available experimental data and preparation predictive models for binding affinity molecules with respect two nuclear receptors involved in endocrine disruption (ED): oestrogen (ER) androgen (AR) receptors. The ED-relevant were retrieved from multiple sources, including CERAPP, CoMPARA, Tox21 projects as well ChEMBL PubChem databases. Data performed help generative topographic mapping revealed problem low agreement between values different...
Abstract The conversion of chemical structures into computer-readable descriptors, able to capture key structural aspects, is pivotal importance in the field cheminformatics and computer-aided drug design. Molecular fingerprints represent a widely employed class descriptors; however, their generation process time-consuming for large databases susceptible bias. Therefore, descriptors accurately detect predefined fragments devoid lengthy procedures would be highly desirable. To meet additional...
Abstract Drugs off-target interactions are one of the main reasons candidate failure in drug discovery process. Anticipating potential drug’s adverse effects early stages is necessary to minimize health risks on patients, animal testing, and economical costs. With constantly increasing size virtual screening libraries AI-driven methods can be exploited as first-tier tools proving liability estimation for candidates. We present ProfhEX, an suite 46 OECD-compliant machine learning models able...