This paper investigates a simple, yet effective method for regression on graphs, in particular for applications in chem-informatics and for quantitative structure-activity relationships (QSARs). The method combines Locally Weighted Learning (LWL) with Maximum Common Subgraph (MCS) based graph distances. More specifically, we investigate a variant of locally weighted regression on graphs (structures) that uses the maximum common subgraph for determining and weighting the neighborhood of a graph and feature vectors for the actual regression model. We show that this combination, LWL-MCS, outperforms other methods that use the local neighborhood of graphs for regression. The performance of this method on graphs suggests it might be useful for other types of structured data as well.

We present a numerical refinement operator based on multi-instance learning. In the approach, the task of handling numerical variables in a clause is delegated to statistical multi-instance learning schemes. To each clause, there is an associated multi-instance classification model with the numerical variables of the clause as input. Clauses are built in a greedy manner, where each refinement adds new numerical variables which are used additionally to the numerical variables already known to the multi-instance model. In our experiments, we tested this approach with multi-instance learners available in the Weka workbench (like MI-SVMs). These clauses are used in a boosting approach that can take advantage of the margin information, going beyond standard covering procedures or the discrete boosting of rules, like in SLIPPER. The approach is evaluated on the problem of hexose binding site prediction, a pharmacological application and mutagenicity prediction. In two of the three applications, the task is to find configurations of points with certain properties in 3D space that characterize either a binding site or drug activity: the logical part of the clause constitutes the points with their properties, whereas the multi-instance model constrains the distances among the points. In summary, the new numerical refinement operator is interesting both theoretically as a new synthesis of logical and statistical learning and practically as a new method for characterizing binding sites and pharmacophores in biochemical applications.

We present a novel (Q)SAR approach that detects groups of structures for local (Q)SAR modeling. The algorithm combines clustering and classification or regression for making predictions on chemical structure data. A clustering procedure producing clusters with shared structural scaffolds is applied as a preprocessing step, before a (local) model is learned for each relevant cluster. Instead of using only one global model (classical approach), we use weighted local models for predictions of query compounds dependent on cluster memberships. The approach is evaluated and compared against standard statistical (Q)SAR algorithms on various datasets. The results show that in many cases the application of local models significantly improves the predictive power of the derived (Q)SAR models compared to the classical approach, to models that are induced by a fingerprint-based or a hierarchical clustering approach and to locally weighted learning.

MOTIVATION:The slow growth of expert-curated databases compared to experimental databases makes it necessary to build upon highly accurate automated processing pipelines to make the most of the data until curation becomes available. We address this problem in the context of protein structures and their classification into structural and functional classes, more specifically, the structural classification of proteins (SCOP). Structural alignment methods like Vorolign already provide good classification results, but effectively work in a 1-Nearest Neighbor mode. Model-based (in contrast to instance-based) approaches so far have been shown to be of limited values due to small classes arising in such classification schemes.RESULTS:In this article, we describe how kernels defined in terms of Vorolign scores can be used in SVM learning, and explore variants of combined instance-based and model-based learning, up to exclusively model-based learning. Our results suggest that kernels based on Vorolign scores are effective and that model-based learning can yield highly competitive classification results for the prediction of SCOP families.AVAILABILITY:The code is made available at: http://wwwkramer.in.tum.de/research/applications/vorolign-kernel.

BACKGROUND:We present a machine learning approach to the problem of protein ligand interaction prediction. We focus on a set of binding data obtained from 113 different protein kinases and 20 inhibitors. It was attained through ATP site-dependent binding competition assays and constitutes the first available dataset of this kind. We extract information about the investigated molecules from various data sources to obtain an informative set of features.RESULTS:A Support Vector Machine (SVM) as well as a decision tree algorithm (C5/See5) is used to learn models based on the available features which in turn can be used for the classification of new kinase-inhibitor pair test instances. We evaluate our approach using different feature sets and parameter settings for the employed classifiers. Moreover, the paper introduces a new way of evaluating predictions in such a setting, where different amounts of information about the binding partners can be assumed to be available for training. Results on an external test set are also provided.CONCLUSIONS:In most of the cases, the presented approach clearly outperforms the baseline methods used for comparison. Experimental results indicate that the applied machine learning methods are able to detect a signal in the data and predict binding affinity to some extent. For SVMs, the binding prediction can be improved significantly by using features that describe the active site of a kinase. For C5, besides diversity in the feature set, alignment scores of conserved regions turned out to be very useful.

Quantitative structure-activity relationships (QSARs) are regression models relating chemical structure to biological activity. Such models allow to make predictions for toxicologically or pharmacologically relevant endpoints, which constitute the target outcomes of trials or experiments. The task is often tackled by instance-based methods (like k-nearest neighbors), which are all based on the notion of chemical (dis-)similarity. Our starting point is the observation by Raymond and Willett that the two big families of chemical distance measures, fingerprint-based and maximum common subgaph based measures, provide orthogonal information about chemical similarity. The paper presents a novel method for finding suitable combinations of them, called adapted transfer, which adapts a distance measure learned on another, related dataset to a given dataset. Adapted transfer thus combines distance learning and transfer learning in a novel manner. In a set of experiments, we compare adapted transfer with distance learning on the target dataset itself and inductive transfer without adaptations. In our experiments, we visualize the performance of the methods by learning curves (i.e., depending on training set size) and present a quantitative comparison for 10% and 100% of the maximum training set size.

OpenTox provides an interoperable, standards-based Framework for the support of predictive toxicology data management, algorithms, modelling, validation and reporting. It is relevant to satisfying the chemical safety assessment requirements of the REACH legislation as it supports access to experimental data, (Quantitative) Structure-Activity Relationship models, and toxicological information through an integrating platform that adheres to regulatory requirements and OECD validation principles. Initial research defined the essential components of the Framework including the approach to data access, schema and management, use of controlled vocabularies and ontologies, architecture, web service and communications protocols, and selection and integration of algorithms for predictive modelling. OpenTox provides end-user oriented tools to non-computational specialists, risk assessors, and toxicological experts in addition to Application Programming Interfaces (APIs) for developers of new applications. OpenTox actively supports public standards for data representation, interfaces, vocabularies and ontologies, Open Source approaches to core platform components, and community-based collaboration approaches, so as to progress system interoperability goals.The OpenTox Framework includes APIs and services for compounds, datasets, features, algorithms, models, ontologies, tasks, validation, and reporting which may be combined into multiple applications satisfying a variety of different user needs. OpenTox applications are based on a set of distributed, interoperable OpenTox API-compliant REST web services. The OpenTox approach to ontology allows for efficient mapping of complementary data coming from different datasets into a unifying structure having a shared terminology and representation.Two initial OpenTox applications are presented as an illustration of the potential impact of OpenTox for high-quality and consistent structure-activity relationship modelling of REACH-relevant endpoints: ToxPredict which predicts and reports on toxicities for endpoints for an input chemical structure, and ToxCreate which builds and validates a predictive toxicity model based on an input toxicology dataset. Because of the extensible nature of the standardised Framework design, barriers of interoperability between applications and content are removed, as the user may combine data, models and validation from multiple sources in a dependable and time-effective way.

Many methods for quantitative structure-activity relationships (QSARs) deliver point estimates only, without quantifying the uncertainty inherent in the prediction. One way to quantify the uncertainy of a QSAR prediction is to predict the conditional density of the activity given the structure instead of a point estimate. If a conditional density estimate is available, it is easy to derive prediction intervals of activities. In this paper, we experimentally evaluate and compare three methods for conditional density estimation for their suitability in QSAR modeling. In contrast to traditional methods for conditional density estimation, they are based on generic machine learning schemes, more specifically, class probability estimators. Our experiments show that a kernel estimator based on class probability estimates from a random forest classifier is highly competitive with Gaussian process regression, while taking only a fraction of the time for training. Therefore, generic machine-learning based methods for conditional density estimation may be a good and fast option for quantifying uncertainty in QSAR modeling.

Automatic classification of proteins using machine learning is an important problem that has received significant attention in the literature. One feature of this problem is that expert-defined hierarchies of protein classes exist and can potentially be exploited to improve classification performance. In this article, we investigate empirically whether this is the case for two such hierarchies. We compare multiclass classification techniques that exploit the information in those class hierarchies and those that do not, using logistic regression, decision trees, bagged decision trees, and support vector machines as the underlying base learners. In particular, we compare hierarchical and flat variants of ensembles of nested dichotomies. The latter have been shown to deliver strong classification performance in multiclass settings. We present experimental results for synthetic, fold recognition, enzyme classification, and remote homology detection data. Our results show that exploiting the class hierarchy improves performance on the synthetic data but not in the case of the protein classification problems. Based on this, we recommend that strong flat multiclass methods be used as a baseline to establish the benefit of exploiting class hierarchies in this area.

This index covers all technical items - papers, correspondence, reviews, etc. - that appeared in this periodical during the year, and items from previous years that were commented upon or corrected in this year. Departments and other items may also be covered if they have been judged to have archival value. The Author Index contains the primary entry for each item, listed under the first author's name. The primary entry includes the co-authors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination. The Subject Index contains entries describing the item under all appropriate subject headings, plus the first author's name, the publication abbreviation, month, and year, and inclusive pages. Note that the item title is found only under the primary entry in the Author Index.

The goal of graph clustering is to partition objects in a graph database into different clusters based on various criteria such as vertex connectivity, neighborhood similarity or the size of the maximum common subgraph. This can serve to structure the graph space and to improve the understanding of the data. In this paper, we present a novel method for structural graph clustering, i.e. graph clustering without generating features or decomposing graphs into parts. In contrast to many related approaches, the method does not rely on computationally expensive maximum common subgraph (MCS) operations or variants thereof, but on frequent subgraph mining. More specifically, our problem formulation takes advantage of the frequent subgraph miner gSpan (that performs well on many practical problems) without effectively generating thousands of subgraphs in the process. In the proposed clustering approach, clusters encompass all graphs that share a sufficiently large common subgraph. The size of the common subgraph of a graph in a cluster has to take at least a user-specified fraction of its overall size. The new algorithm works in an online mode (processing one structure after the other) and produces overlapping (non-disjoint) and nonexhaustive clusters. In a series of experiments, we evaluated the effectiveness and efficiency of the structural clustering algorithm on various real world data sets of molecular graphs.