A5080605090- Computational Drug Discovery Methods
- Spectroscopy and Chemometric Analyses
- Advanced Chemical Sensor Technologies
- Analytical Chemistry and Chromatography
- Metabolomics and Mass Spectrometry Studies
- Advanced Statistical Methods and Models
- Meat and Animal Product Quality
- Fault Detection and Control Systems
- Machine Learning in Materials Science
- History and advancements in chemistry
- Water Quality Monitoring and Analysis
- Fermentation and Sensory Analysis
- Neural Networks and Applications
- Biochemical Analysis and Sensing Techniques
- Chemistry and Chemical Engineering
- Identification and Quantification in Food
- Spectroscopy Techniques in Biomedical and Chemical Research
- Environmental Toxicology and Ecotoxicology
- Analytical chemistry methods development
- Pesticide Residue Analysis and Safety
- Statistical and Computational Modeling
- Geochemistry and Geologic Mapping
- Machine Learning in Bioinformatics
- Analytical Methods in Pharmaceuticals
- Anomaly Detection Techniques and Applications
University of Milano-Bicocca
2016-2025
Data61
2020
Commonwealth Scientific and Industrial Research Organisation
2020
La Trobe University
2020
Monash University
2020
CSIRO Manufacturing
2020
University of Nottingham
2020
Mylan (Switzerland)
2016
Universidade da Coruña
2010
University of Milan
2005-2008
The common steps to calibrate and validate classification models based on partial least squares discriminant analysis are discussed in the present tutorial. All issues be evaluated during model training validation introduced explained using a chemical dataset, composed of toxic non-toxic sediment samples. was carried out with MATLAB routines, which available ESI this tutorial, together dataset detailed list all instructions used for analysis.
One of the OECD principles for model validation requires defining Applicability Domain (AD) QSAR models. This is important since reliable predictions are generally limited to query chemicals structurally similar training compounds used build model. Therefore, characterization interpolation space significant in AD and this study some existing descriptor-based approaches performing task discussed compared by implementing them on validated datasets from literature. Algorithms adopted different...
Abstract This paper deals with the problem of evaluating predictive ability regression models. In some cases, model validation by internal cross‐validation technique is not enough and an external test set has been suggested as effective way ability. Different functions for calculating squared correlation coefficient Q 2 from were proposed, which lead to occasionally different estimates therefore contrasting decisions about adequacy. this paper, advantages drawbacks these in estimating...
Background: Endocrine disrupting chemicals (EDCs) are xenobiotics that mimic the interaction of natural hormones and alter synthesis, transport, or metabolic pathways. The prospect EDCs causing adverse health effects in humans wildlife has led to development scientific regulatory approaches for evaluating bioactivity. This need is being addressed using high-throughput screening (HTS) vitro computational modeling. Objectives: In support Disruptor Screening Program, U.S. Environmental...
The European REACH regulation requires information on ready biodegradation, which is a screening test to assess the biodegradability of chemicals. At same time encourages use alternatives animal testing includes predictions from quantitative structure-activity relationship (QSAR) models. aim this study was build QSAR models predict biodegradation chemicals by using different modeling methods and types molecular descriptors. Particular attention given data validation procedures in order...
la diffusion de documents scientifiques niveau recherche, publiés ou non, émanant des établissements d'enseignement et recherche français étrangers, laboratoires publics privés.
Validation is an essential step of QSAR modeling, and it can be performed by both internal validation techniques (e.g., cross-validation, bootstrap) or external set test objects, that is, objects not used for model development and/or optimization. The evaluation predictive ability then completed comparing experimental predicted values molecules. When dealing with quantitative models, results are generally expressed in terms Q2 metrics. In this work, four fundamental mathematical principles,...
Abstract Natural products are a diverse class of compounds with promising biological properties, such as high potency and excellent selectivity. However, they have different structural motifs than typical drug-like compounds, e.g. , wider range molecular weight, multiple stereocenters higher fraction sp 3 -hybridized carbons. This makes the encoding natural via fingerprints difficult, thus restricting their use in cheminformatics studies. To tackle this issue, we explored over 30 years...
With the growing popularity of using QSAR predictions towards regulatory purposes, such predictive models are now required to be strictly validated, an essential feature which is have model's Applicability Domain (AD) defined clearly. Although in recent years several different approaches been proposed address this goal, no optimal approach define AD has yet recognized.This study proposes a novel descriptor-based method accounts for data distribution and exploits k-Nearest Neighbours (kNN)...
REACH regulation demands information about acute toxicity of chemicals towards fish and supports the use QSAR models, provided compliance with OECD principles. Existing models present some drawbacks that may limit their regulatory application. In this study, a dataset 908 was used to develop model predict LC50 96 hours for fathead minnow. Genetic algorithms combined k nearest neighbour method were applied on training set (726 chemicals) resulted in based six molecular descriptors. An...
In this study, a QSAR model was developed from data set consisting of 546 organic molecules, to predict acute aquatic toxicity toward Daphnia magna. A modified k-Nearest Neighbour (kNN) strategy used as the regression method, which provided prediction only for those molecules with an average distance k nearest neighbours lower than selected threshold. The final showed good performance (R(2) and Q(2) cv equal 0.78, ext 0.72). It comprised eight molecular descriptors that encoded information...