2020
DOI: 10.1002/cem.3223
|View full text |Cite
|
Sign up to set email alerts
|

Mixtures of QSAR models: Learning application domains of pKpredicto rs

Abstract: Quantitative structure‐activity relationship models (QSAR models) predict the physical properties or biological effects based on physicochemical properties or molecular descriptors of chemical structures. Our work focuses on the construction of optimal linear and nonlinear weighted mixes of individual QSAR models to more accurately predict their performance. How the splitting of the application domain by a nonlinear gating network in a “mixture of experts” model structure is suitable for the determination of t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3

Citation Types

0
3
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
3
2

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(3 citation statements)
references
References 42 publications
0
3
0
Order By: Relevance
“…The equation, represented as Eq. 1 , assigns numerical coefficients ( ) to each molecular descriptor, which serve as weighting factors to determine the respective contributions of the predictor variables [ 25 , 81 ]. Although QSAR models are less costly than First Principles, traditional QSAR methods have been hindered by lengthy calculation times, especially when quantum-mechanical electronic descriptors are involved, particularly in large molecules.…”
Section: Introductionmentioning
confidence: 99%
“…The equation, represented as Eq. 1 , assigns numerical coefficients ( ) to each molecular descriptor, which serve as weighting factors to determine the respective contributions of the predictor variables [ 25 , 81 ]. Although QSAR models are less costly than First Principles, traditional QSAR methods have been hindered by lengthy calculation times, especially when quantum-mechanical electronic descriptors are involved, particularly in large molecules.…”
Section: Introductionmentioning
confidence: 99%
“…The equation, represented as Eq. 1, assigns numerical coefficients (a i ) to each molecular descriptor, which serve as weighting factors to determine the respective contributions of the predictor variables [25,81].…”
Section: Introductionmentioning
confidence: 99%
“…The MoE literature for QSAR is very scarce -MoEs were recently investigated (Dörgő et al, 2020) as a QSAR ensemble method across different feature spaces. In methods introduced in the following sections, the problem is naively partitioned by input instances rather than input features, although combining both approaches would be an interesting follow-up.…”
Section: Introductionmentioning
confidence: 99%