Related Experiment Video
Updated: Feb 10, 2026

In Silico Modeling Method for Computational Aquatic Toxicology of Endocrine Disruptors: A Software-Based Approach Using QSAR Toolbox
Published on: August 28, 2019
QSARpy: A new flexible algorithm to generate QSAR models based on dissimilarities. The log Kow case study
Thomas Ferrari1, Anna Lombardo2, Emilio Benfenati1
1IRCCS - Istituto di Ricerche Farmacologiche "Mario Negri", Laboratory of Environmental Chemistry and Toxicology, Via La Masa 19, 20159 Milan, Italy.
QSARpy v1.0 is a novel QSAR modeling tool that uses molecular dissimilarity to predict properties. It accurately estimates the n-octanol/water partition coefficient (log Kow), a key hydrophobicity measure.
Area of Science:
- Computational chemistry
- Quantitative Structure-Activity Relationship (QSAR) modeling
Background:
- Quantitative Structure-Activity Relationship (QSAR) models are crucial for predicting chemical properties and reducing experimental testing.
- Existing automated QSAR development methods rely on atom presence, molecular similarity, or descriptors.
- A novel approach is needed to enhance the accuracy and efficiency of QSAR model development.
Purpose of the Study:
- To introduce QSARpy v1.0, a new QSAR modeling tool utilizing a dissimilarity-based approach.
- To develop and validate a QSAR model for predicting the n-octanol/water partition coefficient (log Kow) using QSARpy v1.0.
- To evaluate the performance of QSARpy v1.0 against existing QSAR modeling tools.
Main Methods:
- QSARpy v1.0 fragments training set molecules to identify 'modulators' – structural fragments associated with property differences.
- The tool predicts a target molecule's property by adjusting training set molecule properties based on shared structures and identified modulators.
- The method was applied to predict the n-octanol/water partition coefficient (log Kow), a measure of hydrophobicity.
Main Results:
- QSARpy v1.0 achieved accurate performance in predicting log Kow.
- The model demonstrated a Root Mean Square Error (RMSE) of 0.43 and an R-squared (R²) of 0.94 on the external test set.
- Performance favorably compared with other existing QSAR modeling programs.
Conclusions:
- QSARpy v1.0 offers a novel and effective dissimilarity-based approach for automated QSAR model development.
- The tool provides accurate predictions for log Kow, a critical parameter in environmental risk assessment.
- QSARpy v1.0 is freely available, promoting wider use in chemical property prediction and cost reduction.
Related Concept Videos
Causes of Similarity-Dissimilarity Effect
The Mantel-Cox Log-Rank Test
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
Trial and Error and Algorithm
Typical Model Studies
¹H NMR of Conformationally Flexible Molecules: Temporal Resolution

