Using support vector machine regression to model the retention of peptides in immobilized metal-affinity chromatography

Kermani, Bahram G.; Kozlov, I. A.; Melnyk, Peter C.; Zhao, Chanfeng; Hachmann, John; Barker, David; Lebl, Michal

doi:10.1016/j.snb.2007.02.004

Cited by 14 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The limit of X outliers is determined by their warning hat values (h*) calculated by 3p/n, where p is the number of model variable plus one (k+1), and n is the number of training compounds. As can be seen from this figure, all predictions were reliable, except for number 7,14,19,191 and 202 in training set which is not within the cut off value of h* = 0.175 and the number of 12, 103, 173 and 212 in William plot seems to be outlier.…”

Section: Mohammad Hossein Fatemi and Fatemeh Fadaeimentioning

confidence: 81%

Quantitative Structure Activity Relationship Prediction of Oral Bioavailabilities Using Support Vector Machine

Fatemi¹,

Fadaei

2014

Journal of the Korean Chemical Society

View full text Add to dashboard Cite

ABSTRACT. A quantitative structure activity relationship (QSAR) study is performed for modeling and prediction of oral bioavailabilities of 216 diverse set of drugs. After calculation and screening of molecular descriptors, linear and nonlinear models were developed by using multiple linear regression (MLR), artificial neural network (ANN), support vector machine (SVM) and random forest (RF) techniques. Comparison between statistical parameters of these models indicates the suitability of SVM over other models. The root mean square errors of SVM model were 5.933 and 4.934 for training and test sets, respectively. Robustness and reliability of the developed SVM model was evaluated by performing of leave many out cross validation test, which produces the statistic of Q SVM 2 = 0.603 and SPRESS = 7.902. Moreover, the chemical applicability domains of model were determined via leverage approach. The results of this study revealed the applicability of QSAR approach by using SVM in prediction of oral bioavailability of drugs.

show abstract

Section: Mohammad Hossein Fatemi and Fatemeh Fadaeimentioning

confidence: 81%

Quantitative Structure Activity Relationship Prediction of Oral Bioavailabilities Using Support Vector Machine

Fatemi¹,

Fadaei

2014

Journal of the Korean Chemical Society

View full text Add to dashboard Cite

show abstract

“…In the case of proteins without a proper affinity partner, alternative technologies include the recombinant fusion of a tag with a highly specific binding partner, such as S-transferase tag [21,22], calmodulin-binding peptide tag [23][24][25], streptavidin tag [26,27], FLAG peptide tag [28,29], and polyhistidine tag [28,30,31]. Recent mathematical modeling related to AC includes a two-zone model (TZM) used for describing the rate of solute uptake by the stationary phase of a sorption-type chromatography [32] and a support vector machine regression model to predict the retention of peptides in immobilized metal-affinity chromatography [33].…”

Section: Affinity Chromatography (Ac)mentioning

confidence: 99%

Mathematical Modeling of Protein Chromatograms

Lienqueo¹,

Mahn²,

Salgado³

et al. 2011

Chem Eng & Technol

View full text Add to dashboard Cite

Artículo de publicación ISIThe most used mathematical models with a phenomenological basis for simulating chromatographic curves of proteins in size exclusion chromatography, ion exchange chromatography, affinity chromatography, and hydrophobic interaction chromatography are reviewed. The plate model (PM) and the general rate model (GRM) are briefly described, followed by various applications of these models to the different chromatographic strategies. Based on these examples it is concluded that the GRM is the most complete and informative model, despite it needs several parameters that have to be estimated from theoretical correlations nonspecific for proteins. Additionally, values for the effective pore diffusion coefficient are not generally available. Appropriate calibration leads in most cases to predictions that compare favorably with experimental data. The possibility to predict chromatographic curves under different operational conditions similar to those used at industrial scale by applying mathematical models is still a challenge because it could contribute to the reduction of costs involved in suitable purification processes. In addition, new proteins are continually designed and for each case different conditions are needed.Fondecyt 1100437 1080143 11080016 Institute for Cell Dynamics and Biotechnology (ICDB): A Center for System Biology P05-001-

show abstract

“…The IMCs of the studied peptides in IMAC were collected from the literature [21]. From the original 306 peptides, 28 compounds were discarded because they had either repeated values or different responding values of IMC to the same peptide.…”

Section: Datasetmentioning

confidence: 99%

“…But there are only few reports of QSRR models for the peptides retention in the IMAC system. Recently, Kermani et al [21] have successfully developed a high-throughput method for the purification of peptide-oligonucleotide conjugates. Double of the three histidines (His3) were placed at the amino-terminus and the 5 0 end of the oligonucleotide, respectively.…”

Section: Introductionmentioning

confidence: 99%

“…Compared with the previous work [21], the dataset used in our investigation is more diverse and the models developed are more reliable and practical. The ultimate aim is to establish a significant model with excellent prediction ability and to obtain some molecular structural information for the interaction retention behavior between the peptides and IMAC.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Novel approaches to predict the retention of histidine‐containing peptides in immobilized metal‐affinity chromatography

Zhang

Wang

et al. 2008

Proteomics

View full text Add to dashboard Cite

The new method lazy learning method-local lazy regression (LLR) was first used to model the quantitative structure-retention relationship (QSRR) for predicting and explaining the retention behaviors of peptides in the nickel column in immobilized metal-affinity chromatography (IMAC). The best multilinear regression (BMLR) method implemented in the CODESSA was used to select the most appropriate molecular descriptors from a large set and build a linear regression model. Based on the selected five descriptors, another two approaches, projection pursuit regression (PPR) and LLR were used to build more accurate QSRR models. The coefficients of determination (R 2 ) of the best model developed based on LLR were 0.9446 and 0.9252 for the training set and the test set, respectively. By comparison, it was proved that the novel local learning method LLR was a very promising tool for QSRR modeling with excellent predictive capability for the prediction of imidazole concentration (IMC) values of histidine-containing peptides in IMAC. It could be used in other chromatography research fields and that should facilitate the design and purification of peptides and proteins.

show abstract

Using support vector machine regression to model the retention of peptides in immobilized metal-affinity chromatography

Cited by 14 publications

References 18 publications

Quantitative Structure Activity Relationship Prediction of Oral Bioavailabilities Using Support Vector Machine

Quantitative Structure Activity Relationship Prediction of Oral Bioavailabilities Using Support Vector Machine

Mathematical Modeling of Protein Chromatograms

Novel approaches to predict the retention of histidine‐containing peptides in immobilized metal‐affinity chromatography

Contact Info

Product

Resources

About