Here a new chapter just published in the Encyclopedia of Analytical Chemistry (Wiley) on distance and simialrity measures for chemical data: Todeschini, R., Ballabio, D. and Consonni, V. (2020). Distances and Similarity Measures in Chemometrics and Chemoinformatics. In Encyclopedia of Analytical Chemistry, R.A. Meyers (Ed.) [link]
New Matlab toolbox available for the calculation of high level data fusion (consensus) approaches (majority voting and Bayes consensus with discrete probability distributions): https://michem.unimib.it/download/data/bayes-and-majority-voting-consensus-for-matlab/
Our latest publication on data fusion, consensus and QSAR is available on line: Valsecchi, C., Grisoni, F., Consonni, V., Ballabio, D. (2020), Consensus versus individual QSARs in classification: comparison on a large-scale case study, Journal of chemical information and modeling, 60, 1215-1223 [link]
This publication was the outcome of a great collaborative project for the identification of Androgen Receptor Activity through machine learning and consensus analysis, check it out: Mansouri, K. et al. (2020) CoMPARA : Collaborative Modeling Project for Androgen Receptor Activity. Environmental Health Perspectives, 128, 027002 [link]
Available PhD position for Molecular Modeling and Virtual Screening for Rational Design of Tubulin-Protein Interaction Modulators at TubInTrain (European Joint Doctorate on chemistry and biology). For further info: https://www.tubintrain.eu/phd-project-esr2/
The valedictory lecture of Prof. Todeschini will take place on monday 16th of december 2019, 10:30, room U3-01 (building U3 – Università degli Studi di Milano – Bicocca): “Una vita per la ricerca, la ricerca di una vita” [link]
In this new release (5.4), class modelling methods (SIMCA, Potential Functions – Kernel Density Estimators, UNEQ) can now be calculated on a specific target class. The toolbox can be downloaded here.
The 23rd EuroQSAR Symposium, entitled “Integrative Data-Intensive Approaches to Drug Design” will take place in Barcelona, September 20-24, 2020: https://www.euroqsar2020.org/
In this new release, batch training has been optimised and now it is faster than previous versions (computational time has been decreased on about 60%). RMSE (squared residuals between sampels and weights of winning neurons) and the average changing of the topological distance between previous and updated winning neurons can be plotted as a function of epochs during training.
The toolbox can be downloaded here.
Durante l’Assemblea dei Soci della Divisione di Chimica Analiticadella Società Chimica Italiana è stata conferita al Prof. Roberto Todeschini la Medaglia Canneri in riconoscimento del suo contributo allo sviluppo e alla divulgazione della chemiometria.