Tanimoto Index Research Articles

Accelerating Two Algorithms for Large-Scale Compound Selection on GPUs

Compound selection procedures based on molecular similarity and diversity are widely used in drug discovery. Current algorithms are often time consuming when applied to very large compound sets. This paper describes the acceleration of two selection algorithms (the leader and the spread algorithms) on graphical processing units (GPUs). We first parallelized the molecular similarity calculation based on Daylight fingerprints and the Tanimoto index and then implemented the two algorithms on GPU hardware using the open source Thrust library. Experiments show that the GPU leader algorithm is 73-120 times faster than the CPU version, and the GPU spread algorithm is 78-143 times faster than the CPU version.

Unsupervised MRI segmentation of brain tissues using a local linear model and level set

Real-world magnetic resonance imaging of the brain is affected by intensity nonuniformity (INU) phenomena which makes it difficult to fully automate the segmentation process. This difficult task is accomplished in this work by using a new method with two original features: (1) each brain tissue class is locally modeled using a local linear region representative, which allows us to account for the INU in an implicit way and to more accurately position the region's boundaries; and (2) the region models are embedded in the level set framework, so that the spatial coherence of the segmentation can be controlled in a natural way. Our new method has been tested on the ground-truthed Internet Brain Segmentation Repository (IBSR) database and gave promising results, with Tanimoto indexes ranging from 0.61 to 0.79 for the classification of the white matter and from 0.72 to 0.84 for the gray matter. To our knowledge, this is the first time a region-based level set model has been used to perform the segmentation of real-world MRI brain scans with convincing results.

Quantitative analysis in clinical applications of brain MRI using independent component analysis coupled with support vector machine

To effectively perform quantification of brain normal tissues and pathologies simultaneously, independent component analysis (ICA) coupled with support vector machine (SVM) is investigated and evaluated for effective volumetric measurements of normal and lesion tissues using multispectral MR images. Synthetic and real MR data of normal brain and white matter lesion (WML) data were used to evaluate the accuracy and reproducibility of gray matter (GM), white matter (WM), and WML volume measurements by using the proposed ICA+SVM method to analyze three sets of MR images, T1-weighted, T2-weighted, and proton density/fluid-attenuated inversion recovery images. The Tanimoto indexes of GM/WM classification in the normal synthetic data calculated by the ICA+SVM method were 0.82/0.89 for data with 0% noise level. As for clinical MR data experiments, the ICA+SVM method clearly extracted the normal tissues and white matter hyperintensity lesions from the MR images, with low intra- and inter-operator coefficient of variations. The experiments conducted provide evidence that the ICA+SVM method has shown promise and potential in applications to classification of normal and pathological tissues in brain MRI.

GPU Accelerated Support Vector Machines for Mining High-Throughput Screening Data

Support Vector Machine (SVM), one of the most promising tools in chemical informatics, is time-consuming for mining large high-throughput screening (HTS) data sets. Here, we describe a parallelization of SVM-light algorithm on a graphic processor unit (GPU), using molecular fingerprints as descriptors and the Tanimoto index as kernel function. Comparison experiments based on six PubChem Bioassay data sets show that the GPU version is 43-104x faster than SVM-light for building classification models and 112-212x over SVM-light for building regression models.

The Poisson Index: a new probabilistic model for protein–ligand binding site similarity

The large-scale comparison of protein-ligand binding sites is problematic, in that measures of structural similarity are difficult to quantify and are not easily understood in terms of statistical similarity that can ultimately be related to structure and function. We present a binding site matching score the Poisson Index (PI) based upon a well-defined statistical model. PI requires only the number of matching atoms between two sites and the size of the two sites-the same information used by the Tanimoto Index (TI), a comparable and widely used measure for molecular similarity. We apply PI and TI to a previously automatically extracted set of binding sites to determine the robustness and usefulness of both scores. We found that PI outperforms TI; moreover, site similarity is poorly defined for TI at values around the 99.5% confidence level for which PI is well defined. A difference map at this confidence level shows that PI gives much more meaningful information than TI. We show individual examples where TI fails to distinguish either a false or a true site paring in contrast to PI, which performs much better. TI cannot handle large or small sites very well, or the comparison of large and small sites, in contrast to PI that is shown to be much more robust. Despite the difficulty of determining a biological 'ground truth' for binding site similarity we conclude that PI is a suitable measure of binding site similarity and could form the basis for a binding site classification scheme comparable to existing protein domain classification schema.

Physicochemical properties/descriptors governing the solubility and partitioning of chemicals in water-solvent-gas systems. Part 2. Solubility in 1-octanol

On the basis of octanol solubility data (log S o) for 218 structurally diverse solid chemicals it was shown that the exclusive consideration of melting points did not provide satisfactory results in the quantitative prediction of this parameter (s = 0.92). The application of HYBOT physicochemical descriptors separately (s = 0.94) and together with melting points (s = 0.70) in the framework of a common regression model also was not successful, although contributions of volume-related and H-bond terms to solubility in octanol were identified. It was proposed that the main reason for such behaviour was the different crystal lattice interaction of different classes of chemicals. Successful calculations of the solubility in octanol of chemicals of interest were performed on the basis of the experimental solubility of structurally/physicochemically/numerically similar nearest neighbours with consideration of their difference in physicochemical parameters (molecular polarisability, H-bond acceptor and donor factors (s = 0.66)) and of these descriptors together with melting point differences (s = 0.38). Good results were obtained for all compounds having nearest neighbours with sufficient similarity, expressed by Tanimoto indexes, and by distances in the scaled 3D descriptor space. Obviously the success of this approach depends on the size of the database.

Similarity measures based on Gaussian‐type promolecular electron density models: Alignment of small rigid molecules

Various molecular similarity measures (overlap, Coulomb, kinetic, electrostatic energy) and similarity indices (Carbó, Hodgkin-Richards, Kulczynski, Shape Tanimoto) are applied to the superposition of 3D promolecular electron density (PED) distributions. The original aspect of the paper lies in the consideration of smoothed PEDs, which allow to decrease the number of local solutions to a superposition problem, together with the use of the less common kinetic and electrostatic energy similarity measures. Results are obtained for a family of five rigid endothiapepsin ligands that were already considered in previous applications, based on graph representations of their PED. In the present work, it is observed that the use of smoothed PED and the kinetic similarity measure, together with the Kulczynski or Shape Tanimoto index, performed the best to align molecules of different sizes.

Multifingerprint Based Similarity Searches for Targeted Class Compound Selection

Molecular fingerprints are widely used for similarity-based virtual screening in drug discovery projects. In this paper we discuss the performance and the complementarity of nine two-dimensional fingerprints (Daylight, Unity, AlFi, Hologram, CATS, TRUST, Molprint 2D, ChemGPS, and ALOGP) in retrieving active molecules by similarity searching against a set of query compounds. For this purpose, we used biological data from HTS screening campaigns of four protein families (GPCRs, kinases, ion channels, and proteases). We have established threshold values for the similarity index (Tanimoto index) to be used as starting points for similarity searches. Based on the complementarities between the selections made by using different fingerprints we propose a multifingerprint approach as an efficient tool to balance the strengths and weaknesses of various fingerprints.

Data-driven brain MRI segmentation supported on edge confidence and a priori tissue information

Brain magnetic resonance imaging segmentation is accomplished in this work by applying nonparametric density estimation, using the mean shift algorithm in the joint spatial-range domain. The quality of the class boundaries is improved by including an edge confidence map, that represents the confidence of truly being in the presence of a border between adjacent regions; an adjacency graph is then constructed with the labeled regions, and analyzed and pruned to merge adjacent regions. In order to assign image regions to a cerebral tissue type, a spatial normalization between image data and standard probability maps is carried out, so that for each structure a maximum a posteriori probability criterion is applied. The method was applied to synthetic and real images, keeping all parameters constant throughout the process for each type of data. The combination of region segmentation and edge detection proved to be a robust technique, as adequate clusters were automatically identified, regardless of the noise level and bias. In a comparison with reference segmentations, average Tanimoto indexes of 0.90-0.99 were obtained for synthetic data and of 0.59-0.99 for real data, considering gray matter, white matter, and background.

Study of spectral analytical data using fingerprints and scaled similarity measurements

A new chemoinformatic model has been developed for enlarging the differences between spectra and applied to differentiation of wines according to the criteria grape origin and variety and ageing process. The model is based on generation of fingerprints from normalised spectra, using empirical parameters and a set of 120 samples. After generation of the fingerprints, similarity matrixes were built on the basis of the Tanimoto similarity index between the fingerprints of the samples. Calculation of the Tanimoto index was modified to adapt the index to the characteristics of the analytical measurements. Thus, scaling factors taking into account pattern fingerprints generated from a group of samples with common characteristics were used. In addition, a modified expression for calculating the Tanimoto index was employed. Principal-components analysis (PCA) and soft independent modelling of class analogy (SIMCA) were applied to the similarity matrixes. The results obtained are discussed as a function of the normalisation method employed, the empirical factor used in generation of the fingerprints, and selection of samples for building the pattern fingerprint, etc. Finally, results from differentiation of wines are compared with those obtained by applying PCA to the unprocessed spectra as stated by the proposed model.

Learnability-based further prediction of gene functions in Gene Ontology

Currently the functional annotations of many genes are not specific enough, limiting their further application in biology and medicine. It is necessary to push the gene functional annotations deeper in Gene Ontology (GO), or to predict further annotated genes with more specific GO terms. A framework of learnability-based further prediction of gene functions in GO is proposed in this paper. Local classifiers are constructed in local classification spaces rooted at qualified parent nodes in GO, and their classification performances are evaluated with the averaged Tanimoto index (ATI). Classification spaces with higher ATIs are selected out, and genes annotated only to the parent classes are predicted to child classes. Through learnability-based further predicting, the functional annotations of annotated genes are made more specific. Experiments on the fibroblast serum response dataset reported further functional predictions for several human genes and also gave interesting clues to the varied learnability between classes of different GO ontologies, different levels, and different numbers of child classes.

Spectral similarity versus structural similarity: mass spectrometry

A recently described method [Anal. Chim. Acta 490 (2003) 313] for the evaluation of spectral similarity searches has been applied to low resolution mass spectra. Aim of the method is to measure the similarity between the chemical structures of query compounds and the chemical structures of the found reference compounds (hits). A high structural similarity is desirable if the query is not present in the spectral library. Similarity of chemical structures has been measured by the Tanimoto index, calculated from 1365 binary substructure descriptors. The method has been applied to sets of 200–10,000 hitlists obtained with different search methods from a database containing 106,955 compounds. Hitlists with highest structure information have been obtained by using a similarity measure based on the correlation coefficient computed either from spectral features, or from the cubic root of peak intensities for masses up to 200. Frequency distributions of spectral and structural similarities have been investigated and a threshold for the spectral similarity has been derived that in general yields hitlists with structures that are very similar to the query structure. An example compares different search methods.

Spectral similarity versus structural similarity: infrared spectroscopy

A new method is described for evaluation of spectral similarity searches. Aim of the method is to measure the similarity between the chemical structures of query compounds and the found reference compounds (hits). A high structural similarity is essential if the query is not present in the spectral library. Similarity of chemical structures was measured by the Tanimoto index, calculated from 1365 binary substructure descriptors. The method has been applied to several 1000 hitlists from searches in an infrared (IR) spectra database containing 13,484 compounds. Hitlists with highest structure information were obtained using a similarity measure based on the correlation coefficient computed from mean centered absorbance units. Frequency distributions of spectral and structural similarities have been investigated and a threshold for the spectral similarity has been derived that in general gives hitlists exhibiting significant chemical structure similarities with the query.

Clustering and similarity of chemical structures represented by binary substructure descriptors

A set of 1365 substructures has been created for the representation of organic chemical structures by binary substructure descriptors. Software SubMat calculates a matrix of binary fingerprint vectors for given sets of molecular structures and substructures (both in Molfile format). The distribution of the substructures in two spectral databases (IR, MS) has been investigated. Examples of structure similarity searches based on the Tanimoto index are presented. Cluster analyses of chemical structures by principal component analysis (PCA) have been performed for hitlists from spectral similarity searches.

An algebraic/graphical tool to compare ecosystems with respect to their pollution by PB/CD III: Comparative regional analysis by applying a similarity index

The technique of Hasse diagrams is applied to compare heavy metal pollution of regions in Germany. Four matrices, i.e. moss, leaves, earthworm ( Lumbricus Rubellus) and herbs are investigated. Beside the pollution pattern of the regions sociogeographic information, i.e. agriculture, industry, woodland proportion and urban zones are given. With regard to this additional information a similarity search among the four different matrices with respect to their loading pattern is presented. The Tanimoto index is applied and it can be shown that low agricultural activity leads to similar Hasse diagrams for the matrices moss-leaves. Therefore a mechanism related to agricultural intensity may be assumed to exist, which leads to a similar pollution pattern for these two matrices.