From owner-chemistry@ccl.net Tue Nov  9 10:25:01 2021
From: "Andrea Mauri andrea.mauri!A!alvascience.com" <owner-chemistry,,server.ccl.net>
To: CCL
Subject: CCL: alvaModel and alvaRunner 2.0
Message-Id: <-54492-211109085544-19954-p8Dyew0Uq04+BDcLxLrang,,server.ccl.net>
X-Original-From: "Andrea  Mauri" <andrea.mauri*o*alvascience.com>
Date: Tue, 9 Nov 2021 08:55:41 -0500


Sent to CCL by: "Andrea  Mauri" [andrea.mauri%x%alvascience.com]
Dear all,
We are happy to inform you that we have just released the version 2.0.0 of
alvaModel and alvaRunner.
alvaModel is a software tool to create Quantitative Structure Activity/Property 
Relationship (QSAR/QSPR) models and alvaRunner is the software tool to apply 
models developed with alvaModel on a new set of molecules. 

Here is the list of the main changes:

Support of classification models:
- Linear and Quadratic Discriminant Analysis (LDA/QDA)
- Partial Least Squares Discriminant Analysis (PLS-DA)
- K-Nearest Neighbors (KNN)
- Support Vector Machine (SVM)
- Consensus model
- added major classification scores (e.g., Sensitivity)

Added new regression models:
- Partial Least Squares (PLS)
- Support Vector Machine (SVM)

Additional features:
- data can be explored using a spreadsheet-like table with sorting and 
filtering capabilities
- external variables can also be imported from CSV and SMILES files
- results can also be exported in SMILES and MDL/SDF file formats
- added new model charts: Williams plot, OLS/PLS beta bars, histograms of 
descriptors and predicted values
- OLS model betas can now be edited manually
- added a functionality to search for the optimal number of neighbours of a KNN 
model
- added a user interface to change the descriptors of a model
- the progress of the Genetic Algorithms is now enriched by a chart showing the 
trend of the score over time
- added a user interface to analyse a single prediction using Atomic 
contributions, Fragment contributions and KNN neighbours (for KNN models)

New versions of alvaRunner and alvaRunner KNIME Plugin have been released to be 
fully compatible with alvaModel v2.0.0. 

Additional information can be found at: https://www.alvascience.com/

Best regards,
Andrea Mauri