1.
Eklund, Martin, et al.
(författare)
The C1C2 : a framework for simultaneous model selection and assessment
2008
Ingår i: BMC Bioinformatics. - : Springer Science and Business Media LLC. - 1471-2105. ; 9, s. 360-
Tidskriftsartikel (refereegranskat) abstract
BACKGROUND: There has been recent concern regarding the inability of predictive modeling approaches to generalize to new data. Some of the problems can be attributed to improper methods for model selection and assessment. Here, we have addressed this issue by introducing a novel and general framework, the C1C2, for simultaneous model selection and assessment. The framework relies on a partitioning of the data in order to separate model choice from model assessment in terms of used data. Since the number of conceivable models in general is vast, it was also of interest to investigate the employment of two automatic search methods, a genetic algorithm and a brute-force method, for model choice. As a demonstration, the C1C2 was applied to simulated and real-world datasets. A penalized linear model was assumed to reasonably approximate the true relation between the dependent and independent variables, thus reducing the model choice problem to a matter of variable selection and choice of penalizing parameter. We also studied the impact of assuming prior knowledge about the number of relevant variables on model choice and generalization error estimates. The results obtained with the C1C2 were compared to those obtained by employing repeated K-fold cross-validation for choosing and assessing a model. RESULTS: The C1C2 framework performed well at finding the true model in terms of choosing the correct variable subset and producing reasonable choices for the penalizing parameter, even in situations when the independent variables were highly correlated and when the number of observations was less than the number of variables. The C1C2 framework was also found to give accurate estimates of the generalization error. Prior information about the number of important independent variables improved the variable subset choice but reduced the accuracy of generalization error estimates. Using the genetic algorithm worsened the model choice but not the generalization error estimates, compared to using the brute-force method. The results obtained with repeated K-fold cross-validation were similar to those produced by the C1C2 in terms of model choice, however a lower accuracy of the generalization error estimates was observed. CONCLUSION: The C1C2 framework was demonstrated to work well for finding the true model within a penalized linear model class and accurately assess its generalization error, even for datasets with many highly correlated independent variables, a low observation-to-variable ratio, and model assumption deviations. A complete separation of the model choice and the model assessment in terms of data used for each task improves the estimates of the generalization error.
2.
Freyhult, Eva, et al.
(författare)
Unbiased descriptor and parameter selection confirms the potential of proteochemometric modelling
2005
Ingår i: BMC Bioinformatics. - : Springer Science and Business Media LLC. - 1471-2105. ; 6, s. 50-
Tidskriftsartikel (refereegranskat) abstract
Background Proteochemometrics is a new methodology that allows prediction of protein function directly from real interaction measurement data without the need of 3D structure information. Several reported proteochemometric models of ligand-receptor interactions have already yielded significant insights into various forms of bio-molecular interactions. The proteochemometric models are multivariate regression models that predict binding affinity for a particular combination of features of the ligand and protein. Although proteochemometric models have already offered interesting results in various studies, no detailed statistical evaluation of their average predictive power has been performed. In particular, variable subset selection performed to date has always relied on using all available examples, a situation also encountered in microarray gene expression data analysis. Results A methodology for an unbiased evaluation of the predictive power of proteochemometric models was implemented and results from applying it to two of the largest proteochemometric data sets yet reported are presented. A double cross-validation loop procedure is used to estimate the expected performance of a given design method. The unbiased performance estimates (P2) obtained for the data sets that we consider confirm that properly designed single proteochemometric models have useful predictive power, but that a standard design based on cross validation may yield models with quite limited performance. The results also show that different commercial software packages employed for the design of proteochemometric models may yield very different and therefore misleading performance estimates. In addition, the differences in the models obtained in the double CV loop indicate that detailed chemical interpretation of a single proteochemometric model is uncertain when data sets are small. Conclusion The double CV loop employed offer unbiased performance estimates about a given proteochemometric modelling procedure, making it possible to identify cases where the proteochemometric design does not result in useful predictive models. Chemical interpretations of single proteochemometric models are uncertain and should instead be based on all the models selected in the double CV loop employed here.
3.
Kontijevskis, Aleksejs, et al.
(författare)
Computational proteomics analysis of HIV-1 protease interactome
2007
Ingår i: Proteins. - : Wiley. - 0887-3585 .- 1097-0134. ; 68:1, s. 305-312
Tidskriftsartikel (refereegranskat) abstract
HIV-1 protease is a small homodimeric enzyme that ensures maturation of HIV virions by cleaving the viral precursor Gag and Gag-Pol polyproteins into structural and functional elements. The cleavage sites in the viral polyproteins share neither sequence homology nor binding motif and the specificity of the HIV-1 protease is therefore only partially understood. Using an extensive data set collected from 16 years of HIV proteome research we have here created a general and predictive rule-based model for HIV-1 protease specificity based on rough sets. We demonstrate that HIV-1 protease specificity is much more complex than previously anticipated, which cannot be defined based solely on the amino acids at the substrate's scissile bond or by any other single substrate amino acid position only. Our results show that the combination of at least three particular amino acids is needed in the substrate for a cleavage event to occur. Only by combining and analyzing massive amounts of HIV proteome data it was possible to discover these novel and general patterns of physico-chemical substrate cleavage determinants. Our study is an example how computational biology methods can advance the understanding of the viral interactomes.
4.
Kontijevskis, Aleksejs, et al.
(författare)
Proteochemometrics mapping of the interaction space for retroviral proteases and their substrates
2009
Ingår i: Bioorganic & Medicinal Chemistry. - : Elsevier BV. - 0968-0896 .- 1464-3391. ; 17:14, s. 5229-5237
Tidskriftsartikel (refereegranskat) abstract
Understanding the complex interactions of retroviral proteases with their ligands is an important scientific challenge in efforts to achieve control of retroviral infections. Development of drug resistance because of high mutation rates and extensive polymorphisms causes major problems in treating the deadly diseases these viruses cause, and prompts efforts to identify new strategies. Here we report a comprehensive analysis of the interaction of 63 retroviral proteases from nine different viral species with their substrates and inhibitors based on publicly available data from the past 17years of retroviral research. By correlating physico-chemical descriptions of retroviral proteases and substrates to their biological activities we constructed a highly statistically valid 'proteochemometric' model for the interactome of retroviral proteases. Analysis of the model indicated amino acid positions in retroviral proteases with the highest influence on ligand activity and revealed general physicochemical properties essential for tight binding of substrates across multiple retroviral proteases. Hexapeptide inhibitors developed based on the discovered general properties effectively inhibited HIV-1 proteases in vitro, and some exhibited uniformly high inhibitory activity against all HIV-1 proteases mutants evaluated. A generalized proteochemometric model for retroviral proteases interactome has been created and analysed in this study. Our results demonstrate the feasibility of using the developed general strategy in the design of inhibitory peptides that can potentially serve as templates for drug resistance-improved HIV retardants.
5.
Kopanchuk, Sergei, et al.
(författare)
Co-operative regulation of ligand binding to melanocortin receptor subtypes : evidence for interacting binding sites
2005
Ingår i: European Journal of Pharmacology. - : Elsevier BV. - 0014-2999 .- 1879-0712. ; 512:2-3, s. 85-95
Tidskriftsartikel (refereegranskat) abstract
This study evaluates the binding the melanocyte stimulating hormone peptide analogue [125I]NDP-MSH to melanocortin receptors MC1, MC3, MC4 and MC5 in insect cell membranes produced by baculovirus expression systems. The presence of Ca2+ was found to be mandatory to achieve specific [125I]NDP-MSH binding to the melanocortin receptors. Although association kinetics of [125I]NDP-MSH followed the regularities of simple bimolecular reactions, the dissociation of [125I]NDP-MSH from the melanocortin receptors was heterogeneous. Eleven linear and cyclic MSH peptides studied displaced the [125I]NDP-MSH binding to the studied melanocortin receptors, with the shapes of their competition curves varying from biphasic or shallow to super-steep (Hill coefficients ranging from 0.4 to 1.5). Notably the same peptide often gave highly different patterns on different melanocortin receptor subtypes; e.g. the MC4 receptor selective antagonist HS131 gave a Hill coefficient of 1.5 on the MC1 receptor but 0.5-0.7 on the MC(3-5) receptors. Adding a mask of one of the peptides to block its high affinity binding did not prevent other competing peptides to yield biphasic competition curves. The data indicate that the binding of MSH peptides to melanocortin receptors are governed by a complex dynamic homotropic co-operative regulations.
6.
Kopanchuk, Sergei, et al.
(författare)
Kinetic evidence for tandemly arranged ligand binding sites in melanocortin 4 receptor complexes
2006
Ingår i: Neurochemistry International. - : Elsevier BV. - 0197-0186 .- 1872-9754. ; 49:5, s. 533-542
Tidskriftsartikel (refereegranskat) abstract
The melanocortin 4 receptor (MC(4)R) binding of the peptide analogue of melanocyte stimulating hormone, [(125)I]NDP-MSH, and the low molecular weight radionucleid 1-(D-1,2,3,4-tetrahydroisoquinoline-3-carboxy-D-4-(125)iodophenylalanyl)-4-cyclohexyl-4-[(1,2,4-triazol-1-yl)methyl]piperidine trifluoroacetate ([(125)I]THIQ) were compared. Kinetic analysis indicated heterogeneity in the binding of both radioligands, the binding apparently proceeding to two tandemly arranged interconnected mutually dependent binding sites. Steric considerations and BRET analysis of Rluc and GFP tagged receptors proposed that these sites are located on different subunits of receptor dimers, which form receptor complexes. According to the minimal model proposed, ligand binding proceeds consecutively to the two binding sites of the dimer. After binding of the first ligand conformational transformations of the complex occur, which is followed by binding of the second ligand. When both receptor units have bound [(125)I]NDP-MSH, the radioligand can be released only from one unit. The [(125)I]NDP-MSH bound to the remaining unit stays practically irreversibly bound due to a very slow retransformation rate of the transformed complex. The considerably faster binding of [(125)I]THIQ did not allow accurate kinetic differentiation of the two binding sites. However, addition of NDP-MSH as well as a fragment of the human agouti protein, hAGRP(83-132) to the preformed [(125)I]THIQ-MC(4)R complex drastically retarded the release of [(125)I]THIQ from the complex, blocking conformational transformations in the complex by binding into the second binding site. The consecutive binding of ligands to the MC(4)R dimers has substantial impact on the apparent ligand potencies, when determined in competition with the two different radioligands applied herein; the apparent potencies of the same ligand differing up to three orders of magnitude when assayed in competition with [(125)I]NDP-MSH or [(125)I]THIQ.
7.
Lapins, Maris, et al.
(författare)
Proteochemometric Modeling of Drug Resistance over the Mutational Space for Multiple HIV Protease Variants and Multiple Protease Inhibitors
2009
Ingår i: Journal of chemical information and modeling. - : American Chemical Society (ACS). - 1549-9596 .- 1549-960X. ; 49:5, s. 1202-1210
Tidskriftsartikel (refereegranskat) abstract
The main therapeutic targets in HIV are its protease and reverse transcriptase. A major problem in treatment of HIV is the ability of the virus to develop drug resistance by accumulating mutations in its targets. Acquiring detailed understanding of the molecular mechanisms for the interactions of drugs with mutated variants of the HIV virus is mandatory to be able to design inhibitors that can evade the resistance. Here we have used proteochemometric modeling to simultaneously analyze the interactions of 21 protease inhibitors with 72 unique protease variants. Inhibition data (pK(i)) were correlated to descriptions of chemical and structural properties of the inhibitors and proteases. The proteochemometric model obtained showed excellent fit and predictive ability (R(2)=0.92, Q(2)=0.83, Q(2)(inh)=0.78) and provided quantitative assessments for the contribution of each mutation and their combinations to the decrease in inhibitor activity, both for the whole compounds series as well as for individual compounds. The model revealed the most deleterious mutations in the protease to be D30N, V32I, G48V, I50V, I54V, V82A, I84V, and L90M. The model was further used to identify molecular properties of chemical compounds that are important for their inhibition of multimutated protease variants. Our results give directions how to design novel improved inhibitors.
8.
Lapins, Maris, et al.
(författare)
Proteochemometric modeling of HIV protease susceptibility
2008
Ingår i: BMC Bioinformatics. - : Springer Science and Business Media LLC. - 1471-2105. ; 9, s. 181-
Tidskriftsartikel (refereegranskat) abstract
BACKGROUNDA major obstacle in treatment of HIV is the ability of the virus to mutate rapidly into drug-resistant variants. A method for predicting the susceptibility of mutated HIV strains to antiviral agents would provide substantial clinical benefit as well as facilitate the development of new candidate drugs. Therefore, we used proteochemometrics to model the susceptibility of HIV to protease inhibitors in current use, utilizing descriptions of the physico-chemical properties of mutated HIV proteases and 3D structural property descriptions for the protease inhibitors. The descriptions were correlated to the susceptibility data of 828 unique HIV protease variants for seven protease inhibitors in current use; the data set comprised 4792 protease-inhibitor combinations.RESULTSThe model provided excellent predictability (R2 = 0.92, Q2 = 0.87) and identified general and specific features of drug resistance. The model's predictive ability was verified by external prediction in which the susceptibilities to each one of the seven inhibitors were omitted from the data set, one inhibitor at a time, and the data for the six remaining compounds were used to create new models. This analysis showed that the over all predictive ability for the omitted inhibitors was Q2 inhibitors = 0.72.CONCLUSIONOur results show that a proteochemometric approach can provide generalized susceptibility predictions for new inhibitors. Our proteochemometric model can directly analyze inhibitor-protease interactions and facilitate treatment selection based on viral genotype. The model is available for public use, and is located at HIV Drug Research Centre.
9.
Lapinsh, Maris, et al.
(författare)
Improved approach for proteochemometrics modeling : application to organic compound--amine G protein-coupled receptor interactions
2005
Ingår i: Bioinformatics. - : Oxford University Press (OUP). - 1367-4803 .- 1367-4811 .- 1460-2059. ; 21:23, s. 4289-4296
Tidskriftsartikel (refereegranskat) abstract
MOTIVATION: Proteochemometrics is a novel technology for the analysis of interactions of series of proteins with series of ligands. We have here customized it for analysis of large datasets and evaluated it for the modeling of the interaction of psychoactive organic amines with all the five known families of amine G protein-coupled receptors (GPCRs). RESULTS: The model exploited data for the binding of 22 compounds to 31 amine GPCRs, correlating chemical descriptions and cross-descriptions of compounds and receptors to binding affinity using a novel strategy. A highly valid model (q2 = 0.76) was obtained which was further validated by external predictions using data for 10 other entirely independent compounds, yielding the high q2ext = 0.67. Interpretation of the model reveals molecular interactions that govern psychoactive organic amines overall affinity for amine GPCRs, as well as their selectivity for particular amine GPCRs. The new modeling procedure allows us to obtain fully interpretable proteochemometrics models using essentially unlimited number of ligand and protein descriptors.
10.
Lapinsh, Maris, et al.
(författare)
Proteochemometric mapping of the interaction of organic compounds with melanocortin receptor subtypes
2005
Ingår i: Molecular Pharmacology. - : American Society for Pharmacology & Experimental Therapeutics (ASPET). - 0026-895X .- 1521-0111. ; 67:1, s. 50-59
Tidskriftsartikel (refereegranskat) abstract
Proteochemometrics was applied in the analysis of the binding of organic compounds to wild-type and chimeric melanocortin receptors. Thirteen chimeric melanocortin receptors were designed based on statistical molecular design; each chimera contained parts from three of the MC(1,3-5) receptors. The binding affinities of 18 compounds were determined for these chimeric melanocortin receptors and the four wild-type melanocortin receptors. The data for 14 of these compounds were correlated to the physicochemical and structural descriptors of compounds, binary descriptors of receptor sequences, and cross-terms derived from ligand and receptor descriptors to obtain a proteochemometric model (correlation was performed using partial least-squares projections to latent structures; PLS). A well fitted mathematical model (R(2) = 0.92) with high predictive ability (Q(2) = 0.79) was obtained. In a further validation of the model, the predictive ability for ligands (Q(2)lig = 0.68) and receptors (Q(2)rec = 0.76) was estimated. The model was moreover validated by external prediction by using the data for the four additional compounds that had not at all been included in the proteochemometric model; the analysis yielded a Q(2)ext = 0.73. An interpretation of the results using PLS coefficients revealed the influence of particular properties of organic compounds on their affinity to melanocortin receptors. Three-dimensional models of melanocortin receptors were also created, and physicochemical properties of the amino acids inside the receptors' transmembrane cavity were correlated to the PLS modeling results. The importance of particular amino acids for selective binding of organic compounds was estimated and used to outline the ligand recognition site in the melanocortin receptors.
