Exploring classification strategies with the CoEPrA 2006 contest.

MOTIVATION: In silico methods to classify compounds as potential drugs that bind to a specific target become increasingly important for drug design. To build classification devices training sets of drugs with known activities are needed. For many such classification problems, not only qualitative but also quantitative information of a specific property (e.g. binding affinity) is available. The latter can be used to build a regression scheme to predict this property for new compounds. Predicting a compound property explicitly is generally more difficult than classifying that the property lies below or above a given threshold value. Hence, an indirect classification that is based on regression may lead to poorer results than a direct classification scheme. In fact, initially researchers are only interested to classify compounds as potential drugs. The activities of these compounds are subsequently measured in wet lab. RESULTS: We propose a novel approach that uses available quantitative information directly for classification rather than first using a regression scheme. It uses a new type of loss function called weighted biased regression. Application of this method to four widely studied datasets of the CoEPrA contest (Comparative Evaluation of Prediction Algorithms, http://coepra.org) shows that it can outperform simple classification methods that do not make use of this additional quantitative information. AVAILABILITY: A stand alone application is available at the webpage http://agknapp.chemie.fu-berlin.de/agknapp/index.php?menu=software&page=PeptideClassifier that can be used to build a model for a peptide training set to be submitted.

Peptide binding at class I major histocompatibility complex scored with linear functions and support vector machines.

We explore two different methods to predict the binding ability of nonapeptides at the class I major histocompatibility complex using a general linear scoring function that defines a separating hyperplane in the feature space of sequences. In absence of suitable data on non-binding nonapeptides we generated sequences randomly from a selected set of proteins from the protein data bank. The parameters of the scoring function were determined by a generalized least square optimization (LSM) and alternatively by the support vector machine (SVM). With the generalized LSM impaired data for learning with a small set of binding peptides and a large set of non-binding peptides can be treated in a balanced way rendering LSM more successful than SVM, while for symmetric data sets SVM has a slight advantage compared to LSM.

Bound Ligand Conformer Revealed by Flexible Structure Alignment in Absence of Crystal Structures: Indirect Drug Design Probed for HIV-1 Protease Inhibitors.

In the absence of structural knowledge on the target protein, the bound ligand conformer (BLC) can be constructed approximately by an indirect drug-design approach that uses a set of ligands binding to the same target. Once the bound ligand conformer (BLC) is known, different strategies of drug design can be pursued. The indirect drug-design approach of the present study is based on the assumption that a set of ligands with chemically different architecture binding to the same target protein carry hidden information of their corresponding true BLCs. It is shown how this information can be extracted by pairwise flexible structure alignment (FSA) using molecular dynamics (MD) simulations with attractive intermolecular interactions that derive from the molecular similarity of the ligands and allow the ligands to adopt the same space. The FSA approach is performed with a newly designed module overlap in the experimental CHARMM-29a1, which soon will become publicly available. Combining the conformations obtained from FSA of different ligand pairs yields consensus ligand conformers (CLCs) that should be similar to the BLCs. This procedure was validated on HIV-1 protease (HIV-P), where at present 44 crystal structures with bound ligands of sufficiently diverse chemical composition are available. The FSA approach identifies four different clusters of HIV-P BLCs. These clusters are consistent with the H-bond patterns of the ligands bound to HIV-P in the crystal structures exhibiting four different binding modes. The cluster-specific CLCs are indeed very similar (rmsd ≈ 2 Å) to the corresponding BLCs from the crystal structure, demonstrating the feasibility of the present approach.