QSPR designer – a program to design and evaluate QSPR models. Case study on pKa prediction
2011
QSPR Designer: A Tool for Predicting pKa
Sample size: 143
publication
Evidence: high
Author Information
Author(s): Skřehota O, Vařeková RS, Geidl S, Kudera M, Sehnal D, Ionescu C-M, Koča J
Primary Institution: National Centre for Biomolecular Research, Masaryk University, Brno, Czech Republic
Hypothesis
Can QSPR models effectively predict pKa values using atomic charges as descriptors?
Conclusion
The QSPR Designer software successfully designed and evaluated 75 models for predicting pKa, achieving a high correlation coefficient.
Supporting Evidence
- The software can read or calculate structural properties of atoms and bonds.
- It allows for the design and parameterization of QSPR models.
- The best model achieved a correlation coefficient of 0.969 for predicting pKa.
Takeaway
The QSPR Designer helps scientists predict how acidic or basic a molecule is by using computer models based on its structure.
Methodology
The software calculates structural properties and evaluates correlations to design QSPR models for predicting pKa.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website