Biomedical subjects

Dimitris K Agrafiotis

Publications and source records attributed to Dimitris K Agrafiotis.

17 recordsLinked to original sources

PubMed2006

A distance geometry heuristic for expanding the range of geometries sampled during conformational search.

A recent study of crystal structures of protein-ligand complexes has shown that bioactive conformations tend to be more extended than random ones (Diller and Merz, J. Comput. Aid. Mol. Des. 2002, 16, 105-112). Existing conformational sampling techniques produce molecular conformations with a distribution of geometric sizes that may not cover that of the bioactive conformations. Here, we describe a simple heuristic for biasing the conformational search toward more extended or compact conformations, while maintaining excellent sampling. The method uses a boosting strategy to generate a series of conformations, each of which is at least as extended (or compact) as the previous one. We demonstrate that this method significantly expands the range of geometric sizes generated during the search and thus increases the efficiency of sampling bioactive conformations.

Algorithms↗

PubMed2006

A QSAR model of HERG binding using a large, diverse, and internally consistent training set.

Over the past decade, the pharmaceutical industry has begun to address an addition to ADME/Tox profiling--the ability of a compound to bind to and inhibit the human ether-a-go-go-related gene (hERG)-encoded cardiac potassium channel. With the compilation of a large and diverse set of compounds measured in a single, consistent hERG channel inhibition assay, we recognized a unique opportunity to attempt to construct predictive QSAR models. Early efforts with classification models built from this training set were very encouraging. Here, we report a systematic evaluation of regression models based on neural network ensembles in conjunction with a variety of structure representations and feature selection algorithms. The combination of these modeling techniques (neural networks to capture non-linear relationships in the data, feature selection to prevent over-fitting, and aggregation to minimize model instability) was found to produce models with very good internal cross-validation statistics and good predictivity on external data.

Algorithms↗

PubMed2004

A method for quantifying and visualizing the diversity of QSAR models.

Feature selection is one of the most commonly used and reliable methods for deriving predictive quantitative structure-activity relationships (QSAR). Many feature selection algorithms are stochastic in nature and often produce different solutions depending on the initialization conditions. Because some features may be highly correlated, models that are based on different sets of descriptors may capture essentially the same information, however, such models are difficult to recognize. Here, we introduce a measure of similarity between QSAR models that captures the correlation between the underlying features. This measure can be used in conjunction with stochastic proximity embedding (SPE) or multi-dimensional scaling (MDS) to create a meaningful visual representation of structure-activity model space and aid in the post-processing and analysis of results of feature selection calculations.

Algorithms↗

PubMed2003

Stochastic proximity embedding.

We introduce stochastic proximity embedding (SPE), a novel self-organizing algorithm for producing meaningful underlying dimensions from proximity data. SPE attempts to generate low-dimensional Euclidean embeddings that best preserve the similarities between a set of related observations. The method starts with an initial configuration, and iteratively refines it by repeatedly selecting pairs of objects at random, and adjusting their coordinates so that their distances on the map match more closely their respective proximities. The magnitude of these adjustments is controlled by a learning rate parameter, which decreases during the course of the simulation to avoid oscillatory behavior. Unlike classical multidimensional scaling (MDS) and nonlinear mapping (NLM), SPE scales linearly with respect to sample size, and can be applied to very large data sets that are intractable by conventional embedding procedures. The method is programmatically simple, robust, and convergent, and can be applied to a wide range of scientific problems involving exploratory data analysis and visualization.

Journal Article↗

PubMed2003

A modified update rule for stochastic proximity embedding.

Recently, we described a fast self-organizing algorithm for embedding a set of objects into a low-dimensional Euclidean space in a way that preserves the intrinsic dimensionality and metric structure of the data [Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 15869-15872]. The method, called stochastic proximity embedding (SPE), attempts to preserve the geodesic distances between the embedded objects, and scales linearly with the size of the data set. SPE starts with an initial configuration, and iteratively refines it by repeatedly selecting pairs of objects at random, and adjusting their coordinates so that their distances on the map match more closely their respective proximities. Here, we describe an alternative update rule that drastically reduces the number of calls to the random number generator and thus improves the efficiency of the algorithm.

Algorithms↗

PubMed2003

Exploring the nonlinear geometry of protein homology.

The explosion of biological data resulting from genomic and proteomic research has created a pressing need for data analysis techniques that work effectively on a large scale. An area of particular interest is the organization and visualization of large families of protein sequences. An increasingly popular approach is to embed the sequences into a low-dimensional Euclidean space in a way that preserves some predefined measure of sequence similarity. This method has been shown to produce maps that exhibit global order and continuity and reveal important evolutionary, structural, and functional relationships between the embedded proteins. However, protein sequences are related by evolutionary pathways that exhibit highly nonlinear geometry, which is invisible to classical embedding procedures such as multidimensional scaling (MDS) and nonlinear mapping (NLM). Here, we describe the use of stochastic proximity embedding (SPE) for producing Euclidean maps that preserve the intrinsic dimensionality and metric structure of the data. SPE extends previous approaches in two important ways: (1) It preserves only local relationships between closely related sequences, thus allowing the map to unfold and reveal its intrinsic dimension, and (2) it scales linearly with the number of sequences and therefore can be applied to very large protein families. The merits of the algorithm are illustrated using examples from the protein kinase and nuclear hormone receptor superfamilies.

Algorithms↗

PubMed2002

A self-organizing principle for learning nonlinear manifolds.

Modern science confronts us with massive amounts of data: expression profiles of thousands of human genes, multimedia documents, subjective judgments on consumer products or political candidates, trade indices, global climate patterns, etc. These data are often highly structured, but that structure is hidden in a complex set of relationships or high-dimensional abstractions. Here we present a self-organizing algorithm for embedding a set of related observations into a low-dimensional space that preserves the intrinsic dimensionality and metric structure of the data. The embedding is carried out by using an iterative pairwise refinement strategy that attempts to preserve local geometry while maintaining a minimum separation between distant objects. In effect, the method views the proximities between remote objects as lower bounds of their true geodesic distances and uses them as a means to impose global structure. Unlike previous approaches, our method can reveal the underlying geometry of the manifold without intensive nearest-neighbor or shortest-path computations and can reproduce the true geodesic distances of the data points in the low-dimensional embedding without requiring that these distances be estimated from the data sample. More importantly, the method is found to scale linearly with the number of points and can be applied to very large data sets that are intractable by conventional embedding procedures.

Journal Article↗

PubMed2002

Feature selection for structure-activity correlation using binary particle swarms.

We present a new feature selection algorithm for structure-activity and structure-property correlation based on particle swarms. Particle swarms explore the search space through a population of individuals that adapt by returning stochastically toward previously successful regions, influenced by the success of their neighbors. This method, which was originally intended for searching multidimensional continuous spaces, is adapted to the problem of feature selection by viewing the location vectors of the particles as probabilities and employing roulette wheel selection to construct candidate subsets. The algorithm is applied in the construction of parsimonious quantitative structure-activity relationship (QSAR) models based on feed-forward neural networks and is tested on three classical data sets from the QSAR literature. It is shown that the method compares favorably with simulated annealing and is able to identify a better and more diverse set of solutions given the same amount of simulation time.

Algorithms↗

PubMed2002

Combinatorial informatics in the post-genomics ERA.

The multitude of potential drug targets emerging from genome sequencing demands new approaches to drug discovery. A chemogenomics strategy, which involves the generation of small-molecule compounds that can be used both as tools to probe biological mechanisms and as leads for drug-property optimization, provides a highly parallel, industrialized solution. Key to the success of this strategy is an integrated suite of chemi-informatics applications that can allow the rapid and directed optimization of chemical compounds with drug-like properties using 'just-in-time' combinatorial chemical synthesis. An effective embodiment of this process requires new computational and data-mining tools that cover all aspects of library generation, compound selection and experimental design, and work effectively on a massive scale.

Combinatorial Chemistry Techniques↗

PubMed2002

Scalable methods for the construction and analysis of virtual combinatorial libraries.

One can distinguish between two kinds of virtual combinatorial libraries: viable and accessible . Viable libraries are relatively small in size, are assembled from readily available reagents that have been filtered by the medicinal chemist, and often have a physical counterpart. Conversely, accessible libraries can encompass millions or billions of structures, typically include all possible reagents that are in principle compatible with a particular reaction scheme, and they can never be physically synthesized in their entirety. Although the analysis of viable virtual libraries is relatively straightforward, the handling of large accessible libraries requires methods that scale well with respect to library size. In this work, we present novel, efficient and scalable techniques for the construction, analysis, and in silico screening of massive virtual combinatorial libraries.

Combinatorial Chemistry Techniques↗

PubMedDate not supplied

A fractal approach for selecting an appropriate bin size for cell-based diversity estimation.

A novel approach for selecting an appropriate bin size for cell-based diversity assessment is presented. The method measures the sensitivity of the diversity index as a function of grid resolution, using a box-counting algorithm that is reminiscent of those used in fractal analysis. It is shown that the relative variance of the diversity score (sum of squared cell occupancies) of several commonly used molecular descriptor sets exhibits a bell-shaped distribution, whose exact characteristics depend on the distribution of the data set, the number of points considered, and the dimensionality of the feature space. The peak of this distribution represents the optimal bin size for a given data set and sample size. Although box counting can be performed in an algorithmically efficient manner, the ability of cell-based methods to distinguish between subsets of different spread falls sharply with dimensionality, and the method becomes useless beyond a few dimensions.

Journal Article↗

PubMedDate not supplied

On the use of neural network ensembles in QSAR and QSPR.

Despite their growing popularity among neural network practitioners, ensemble methods have not been widely adopted in structure-activity and structure-property correlation. Neural networks are inherently unstable, in that small changes in the training set and/or training parameters can lead to large changes in their generalization performance. Recent research has shown that by capitalizing on the diversity of the individual models, ensemble techniques can minimize uncertainty and produce more stable and accurate predictors. In this work, we present a critical assessment of the most common ensemble technique known as bootstrap aggregation, or bagging, as applied to QSAR and QSPR. Although aggregation does offer definitive advantages, we demonstrate that bagging may not be the best possible choice and that simpler techniques such as retraining with the full sample can often produce superior results. These findings are rationalized using Krogh and Vedelsby's decomposition of the generalization error into a term that measures the average generalization performance of the individual networks and a term that measures the diversity among them. For networks that are designed to resist over-fitting, the benefits of aggregation are clear but not overwhelming.

Computer Simulation↗

PubMedDate not supplied

A geodesic framework for analyzing molecular similarities.

A fast self-organizing algorithm for extracting the minimum number of independent variables that can fully describe a set of observations was recently described (Agrafiotis, D. K.; Xu, H. Proc. Natl. Acad. Sci.U.S.A. 2002, 99, 15869-15872). The method, called stochastic proximity embedding (SPE), attempts to generate low-dimensional Euclidean maps that best preserve the similarities between a set of related objects. Unlike conventional multidimensional scaling (MDS) and nonlinear mapping (NLM), SPE preserves only local relationships and, by doing so, reveals the intrinsic dimensionality and metric structure of the data. Its success depends critically on the choice of the neighborhood radius, which should be consistent with the local curvature of the underlying manifold. Here, we describe a procedure for determining that radius by examining the tradeoff between the stress function and the number of connected components in the neighborhood graph and show that it can be used to produce meaningful maps in any embedding dimension. The power of the algorithm is illustrated in two major areas of computational drug design: conformational analysis and diversity profiling of large chemical libraries.

Journal Article↗

PubMedDate not supplied

Conformational sampling by self-organization.

A new stochastic algorithm for conformational sampling is described. The algorithm generates molecular conformations that are consistent with a set of geometric constraints, which include interatomic distance bounds and chiral volumes derived from the molecular connectivity table. The algorithm repeatedly selects individual geometric constraints at random and updates the respective atomic coordinates toward satisfying the chosen constraint. When compared to a conventional distance geometry algorithm based on the same set of geometric constraints, our method is faster and generates conformations that are more diverse and more energetically favorable.

Algorithms↗

PubMedDate not supplied

Nearest neighbor search in general metric spaces using a tree data structure with a simple heuristic.

We present a new algorithm for nearest neighbor search in general metric spaces. The algorithm organizes the database into recursively partitioned Voronoi regions and represents these partitions in a tree. The separations between the Voronoi regions as well as the radius of each region are used with triangular inequality to derive the minimum possible distance between any point in a region and the query and to discard the region from further search if a smaller distance has already been found. The algorithm also orders the search sequence of the tree branches using the estimate of the minimum possible distance. This simple heuristic proves to considerably enhance the pruning of the search tree. The efficiency of the algorithm is demonstrated on several artificial data sets and real problems in computational chemistry.

Journal Article↗

PubMedDate not supplied

A cluster-based strategy for assessing the overlap between large chemical libraries and its application to a recent acquisition.

We report on the structural comparison of the corporate collections of Johnson & Johnson Pharmaceutical Research & Development (JNJPRD) and 3-Dimensional Pharmaceuticals (3DP), performed in the context of the recent acquisition of 3DP by JNJPRD. The main objective of the study was to assess the druglikeness of the 3DP library and the extent to which it enriched the chemical diversity of the JNJPRD corporate collection. The two databases, at the time of acquisition, collectively contained more than 1.1 million compounds with a clearly defined structural description. The analysis was based on a clustering approach and aimed at providing an intuitive quantitative estimate and visual representation of this enrichment. A novel hierarchical clustering algorithm called divisive k-means was employed in combination with Kelley's cluster-level selection method to partition the combined data set into clusters, and the diversity contribution of each library was evaluated as a function of the relative occupancy of these clusters. Typical 3DP chemotypes enriching the diversity of the JNJPRD collection were catalogued and visualized using a modified maximum common substructure algorithm. The joint collection of JNJPRD and 3DP compounds was also compared to other databases of known medicinally active or druglike compounds. The potential of the methodology for the analysis of very large chemical databases is discussed.

Algorithms↗

PubMedDate not supplied

Using particle swarms for the development of QSAR models based on K-nearest neighbor and kernel regression.

We describe the application of particle swarms for the development of quantitative structure-activity relationship (QSAR) models based on k-nearest neighbor and kernel regression. Particle swarms is a population-based stochastic search method based on the principles of social interaction. Each individual explores the feature space guided by its previous success and that of its neighbors. Success is measured using leave-one-out (LOO) cross validation on the resulting model as determined by k-nearest neighbor kernel regression. The technique is shown to compare favorably to simulated annealing using three classical data sets from the QSAR literature.

Algorithms↗