On further application of <i>r</i> as a metric for validation of QSAR models

Mitra, Indrajit; Roy, Partha Pratim; Kar, Supratik; Ojha, Probir Kumar; Roy, Kunal

doi:10.1002/cem.1268

Cited by 83 publications

(44 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Initially, the concept leading to r 2 m was applied only to the test set prediction [42], but it can also be applied for the training set if one considers the correlation between observed and LOO-predicted values of the training set compounds [43,44]. More interestingly, this can be used for the whole set considering LOO-predicted values for the training set and SAR and QSAR in Environmental Research 155 predicted values of the test set compounds.…”

mentioning

confidence: 99%

QSAR with quantum topological molecular similarity indices: toxicity of aromatic aldehydes toTetrahymena pyriformis

Kar

Harding

Roy

et al. 2010

SAR and QSAR in Environmental Research

View full text Add to dashboard Cite

Extensive production and utilization of aromatic aldehydes and their derivatives without proper certification is alarming with regard to environmental safety. This concern motivated our construction of predictive quantitative structure-activity relationship (QSAR) models for the toxicity of aldehydes to the ecologically important species Tetrahymena pyriformis. Quantum topological molecular similarity (QTMS) descriptors, along with the lipid-water partition coefficient (log K(o/w)), were used as predictor variables. The QTMS descriptors were calculated at different levels of theory including AM1, HF/3-21G(d), HF/6-31G(d), B3LYP/6-31 + G(d,p), B3LYP/6-311 + G(2d,p) and MP2/6-311+G(2d,p). The data set of 77 aromatic aldehydes was divided into a training set (n = 58) and a test (n = 19) set, and 58 models were developed using partial least squares (PLS) and genetic partial least squares (G/PLS). We evaluated the overall predictive capacity of the models based on leave-one-out predictions for the training set compounds and model derived predictions for the test set compounds. For both PLS and G/PLS, the models built at the HF/6-31G(d) level show better predictivity (based on overall prediction) than the models developed at any of the other five levels. Further validation was also performed utilizing (process and model) randomization tests. We show that improved predictive QSAR models for aldehydic toxicity to Tetrahymena pyriformis can be generated using QTMS descriptors along with log K(o/w).

show abstract

mentioning

confidence: 99%

QSAR with quantum topological molecular similarity indices: toxicity of aromatic aldehydes toTetrahymena pyriformis

Kar

Harding

Roy

et al. 2010

SAR and QSAR in Environmental Research

View full text Add to dashboard Cite

show abstract

“…2 Steps associated with the different QSAR techniques performed in the present study map the test set compounds, and the activity of the test molecules was estimated based on the degree of mapping and the calculated value of the external predictive parameter (R 2 pred ; with a threshold value of 0.5) [34]. Furthermore, to better determine the external predictive potential of the developed 3D pharmacophore model, the value of modified r 2 for the test set r 2 m test ð Þ h i was also calculated [35][36][37][38]. Another method employed to determine the essential structural features of the chromone derivatives include the CoMSIA technique [39].…”

Section: Different Methods Employed For the Development Of Qsar Modelsmentioning

confidence: 99%

Development of multiple QSAR models for consensus predictions and unified mechanistic interpretations of the free-radical scavenging activities of chromone derivatives

2011

Self Cite

View full text Add to dashboard Cite

Antioxidants are important defenders of the human body against nocive free radicals, which are the causative agents of most life-threatening diseases. The immense biomedicinal utility of antioxidants necessitates the development and design of new synthetic antioxidant molecules. The present report deals with the modeling of a series of chromone derivatives, which was done to provide detailed insight into the main structural fragments that impart antioxidant activity to these molecules. Four different quantitative structure-property relationship (QSAR) techniques, namely 3D pharmacophore mapping, comparative molecular similarity indices analysis (CoMSIA 3D-QSAR), hologram QSAR (HQSAR), and group-based QSAR (G-QSAR) techniques, were employed to obtain statistically significant models with encouraging external predictive potentials. Moreover, the visual contribution maps obtained for the different models signify the importance of different structural features in specific regions of the chromone nucleus. Additionally, the G-QSAR models determine the composite influence of pairs of substituent fragments on the overall antioxidant activity profiles of the molecules. Multiple models with different strategies for assessing structure-activity relationships were applied to reach a unified conclusion regarding the antioxidant mechanism and to provide consensus predictions, which are more reliable than values derived from a single model. The structural information obtained from the various QSAR models developed in the present work can thus be effectively utilized to design and predict the activities of new molecules belonging to the class of chromone derivatives.

show abstract

“…The slope of predicted versus experimental values of the data through the origin (K) should also lie between 0.85 and 1.15. Recently, a parameter named as 'R 2 m ' which can be calculated by Equation (10) is introduced by Mitra et al [45] for the validation of the model:…”

Section: Qsar Modellingmentioning

confidence: 99%

The use of ladder particle swarm optimisation for quantitative structure–activity relationship analysis of human immunodeficiency virus-1 integrase inhibitors

Jalali‐Heravi

Ebrahimi‐Najafabadi

2011

Molecular Simulation

View full text Add to dashboard Cite

This contribution focuses on the use of ladder particle swarm optimisation (LPSO) on modelling of oxadiazole-and triazolesubstituted naphthyridines as human immunodeficiency virus-1 integrase inhibitors. Artificial neural network (ANN) and Monte Carlo cross-validation techniques were combined with LPSO to develop a quantitative structure -activity relationship model. The techniques of LPSO, ANN and sample set partitioning based on joint x -y distances were applied as feature selection, mapping and model evaluation, respectively. The variables selected by LPSO were used as inputs of Bayesian regularisation ANN. The statistical parameters of correlation of deterministic, R 2 , and root-mean-square error for the test set were 0.876 and 0.23, respectively. Robustness of the model was confirmed by Y-randomisation method. Comparison of the LPSO-ANN results with those of stepwise multiple linear regression (MLR), LPSO-MLR and LPSO-MLR-ANN showed the superiority of LPSO-ANN. Inspection of the selected variables indicated that atomic mass, molecular size and electronic structure of the molecules play a significant role in inhibitory behaviour of oxadiazole-and triazole-substituted naphthyridines.

show abstract

On further application of r as a metric for validation of QSAR models

Cited by 83 publications

References 72 publications

QSAR with quantum topological molecular similarity indices: toxicity of aromatic aldehydes toTetrahymena pyriformis

QSAR with quantum topological molecular similarity indices: toxicity of aromatic aldehydes toTetrahymena pyriformis

Development of multiple QSAR models for consensus predictions and unified mechanistic interpretations of the free-radical scavenging activities of chromone derivatives

The use of ladder particle swarm optimisation for quantitative structure–activity relationship analysis of human immunodeficiency virus-1 integrase inhibitors

Contact Info

Product

Resources

About