How to rank and discriminate artificial neural networks? Case study: prediction of anticancer activity of 17-picolyl and 17-picolinylidene androstane derivatives

Kоvаčеvić, Strаhinjа; Podunavac‐Kuzmanović, Sanja O.; Jevrić, Lidija R.; Djurendić, Evgenija A.; Ajduković, Jovana J.; Gadžurić, Slobodan

doi:10.1007/s13738-015-0759-9

Cited by 11 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SRD methodology, its algorithms and practical uses are described in details in literature. [24][25][26] The validation of SRD analysis was done by comparison of ranks by random numbers (CRRN) and 7-fold cross-validation. 24…”

Section: Chemometric Methodsmentioning

confidence: 99%

Electrostatic and Topological Features as Predictors of Antifungal Potential of Oxazolo Derivatives as Promising Compounds in Treatment of Infections Caused by Candida albicans

Kоvаčеvić

Karadžić

Podunavac‐Kuzmanović

et al. 2018

ACSi

View full text Add to dashboard Cite

The results presented in this study include the prediction of the antifungal activity of 24 oxazolo derivatives based on their topological and electrostatic molecular descriptors, derived from the 2D molecular structures. The artificial neural network (ANN) method was applied as a regression tool. The input data for ANN modeling were selected by stepwise selection (SS) procedure. The ANN modeling resulted in three networks with the outstanding statistical characteristics. High predictivity of the established networks was confirmed by comparisons of the predicted and experimental data and by the residuals analysis. The obtained results indicate the usefulness of the formed ANNs in precise prediction of minimum inhibitory concentrations of the analyzed compounds towards Candida albicans. The Sum of Ranking Differences (SRD) method was used in this study to reveal possible grouping of the compounds in the space of the variables used in ANN modeling. The obtained results can be considered to be a contribution to development of new antifungal drugs structurally based on oxazole core, particularly nowadays when there is a lack of highly efficient antimycotics.

show abstract

Section: Chemometric Methodsmentioning

confidence: 99%

Electrostatic and Topological Features as Predictors of Antifungal Potential of Oxazolo Derivatives as Promising Compounds in Treatment of Infections Caused by Candida albicans

Kоvаčеvić

Karadžić

Podunavac‐Kuzmanović

et al. 2018

ACSi

View full text Add to dashboard Cite

show abstract

“…The sum of ranking differences (SRD) is a recently proposed method to determine the similarity between models and facilitate the selection of models based on their own evaluation merits without considering weight allocation problems . Accordingly, the SRD method was employed to objectively select the optimal parameter combination of the latent variable and the threshold value of the variable importance index in this work.…”

Section: Introductionmentioning

confidence: 99%

“…The sum of ranking differences (SRD) [24][25][26] is a recently proposed method to determine the similarity between models and facilitate the selection of models based on their own evaluation merits without considering weight allocation problems. [27][28][29] Accordingly, the SRD method was employed to objectively select the optimal parameter combination of the latent variable and the threshold value of the variable importance index in this work. The main implementation of the SRD process is as follows: first, all possible parameter combinations of latent variables and threshold values of the variable importance index of corresponding models were input into the SRD input matrix as columns, and several merits reflecting the model bias or variance were input as rows.…”

mentioning

confidence: 99%

Tuning parameter identification for variable selection algorithm using the sum of ranking differences algorithm

Nie

Meng

Chen

et al. 2019

Journal of Chemometrics

View full text Add to dashboard Cite

Variable selection algorithms are often adopted to select the optimal variable from a full set of variables and are efficient for reducing the variable dimension and improving the model accuracy. Nonetheless, the parameters of the variable selection method and regression model, such as the number of latent variables of the partial least squares (PLS) model and the threshold value of the variable importance index, need to be identified. The parameters directly determine the final performance of the model. Currently, these parameters are often determined subjectively. As a result, the model results may be accidental because of the subjective determination of the parameters. To objectively identify these parameters, the sum of ranking differences (SRD) coupled with partial least squares‐variable importance in projection (PLS‐VIP‐SRD) and partial least squares‐uninformative variable elimination (PLS‐UVE‐SRD) algorithms was applied to determine the latent variable of the PLS model and the threshold value of the variable importance index. Furthermore, public near‐infrared data of corn were used as the calculation data. The final results show that the PLS‐VIP‐SRD and PLS‐UVE‐SRD models can more effectively and objectively determine the optimal parameter combination than the PLS‐VIP and PLS‐UVE models. Moreover, the selected variables are easier to interpret, and the prediction accuracy is also improved to some extent.

show abstract

“…However, recently, the sum of ranking differences (SRD) has emerged in the field of multiobjective optimization (MOO) as a statistically sound and simple method for ranking of alternative solutions based on multiple criteria. So far, the method has been applied for fusion of several outlier measures in order to enhance the selection of spectral outliers, merging agrochemical parameters for selection of multiperspective crops and products, fusion of multiple lipophilicity measures for ranking thiepine derivatives, and combining multiple model performance parameters for selection of the optimal models . Although Rácz et al have reported good agreement of the SRD and Derringer desirability approach, Csambalik et al have pointed out that initial data pretreatment, especially criteria scaling, may strongly influence the outcomes of the SRD analysis.…”

Section: Introductionmentioning

confidence: 99%

“…So far, the method has been applied for fusion of several outlier measures in order to enhance the selection of spectral outliers, 13 merging agrochemical parameters for selection of multiperspective crops and products, [14][15][16] fusion of multiple lipophilicity measures for ranking thiepine derivatives, 17 and combining multiple model performance parameters for selection of the optimal models. 18,19 Although Rácz et al 19 have reported good agreement of the SRD and Derringer desirability approach, 20 Csambalik et al 14 have pointed out that initial data pretreatment, especially criteria scaling, may strongly influence the outcomes of the SRD analysis. However, despite its increasing use, there is still no comprehensive evaluation of the SRD applications in MOO and multicriteria decision making (MCDM) and no consensus or guidance on how to properly perform the SRD analysis in this context.…”

mentioning

confidence: 99%

Towards polypotent natural products: The Derringer desirability approach and nonparametric ranking for multicriteria evaluation of essential oils

Andrić

2018

Journal of Chemometrics

View full text Add to dashboard Cite

Chemical complexity of natural products often results in their pharmacological polypotency. However, selecting a natural product with desirable activity profile is not a straightforward task, especially if optimization of one feature results in deterioration of other facets. Recently, in the field of multiobjective optimization, the sum of ranking differences (SRD) has emerged as a simple and statistically sound method for fusion of multiple criteria. However, the data pretreatment seems to strongly influence the ranking outcome, which may lead to ambiguous or even false interpretations. Therefore, in the present study, the data of 55 essential oils originated from different plant species and tested on multiple bacterial and fungal strains as well as 4 antioxidative assays were studied. Essential oils were ranked using the classical Deringer desirability approach, and results were compared with the SRD analysis of primary activity data as well as the previously row‐wise standardized data, normalized data, and the data scaled to fit the preferences. Ultimately, the most promising candidate (polypotent) essential oils, as well as the most resilient and most sensitive bacterial and fungal strains, and antioxidative assays were identified. Data transformation based on the Deringer desirability approach, compared with the data that have not been previously transformed or those that transformed using a row‐wise standardization or normalization to the unit length vector, seemed to be the crucial step providing the sound and meaningful SRD ranking.

show abstract

How to rank and discriminate artificial neural networks? Case study: prediction of anticancer activity of 17-picolyl and 17-picolinylidene androstane derivatives

Cited by 11 publications

References 18 publications

Electrostatic and Topological Features as Predictors of Antifungal Potential of Oxazolo Derivatives as Promising Compounds in Treatment of Infections Caused by Candida albicans

Electrostatic and Topological Features as Predictors of Antifungal Potential of Oxazolo Derivatives as Promising Compounds in Treatment of Infections Caused by Candida albicans

Tuning parameter identification for variable selection algorithm using the sum of ranking differences algorithm

Towards polypotent natural products: The Derringer desirability approach and nonparametric ranking for multicriteria evaluation of essential oils

Contact Info

Product

Resources

About