Distribution Patterns of Odonate Assemblages in Relation to Environmental Variables in Streams of South Korea

Odonata species are sensitive to environmental changes, particularly those caused by humans, and provide valuable ecosystem services as intermediate predators in food webs. We aimed: (i) to investigate the distribution patterns of Odonata in streams on a nationwide scale across South Korea; (ii) to evaluate the relationships between the distribution patterns of odonates and their environmental conditions; and (iii) to identify indicator species and the most significant environmental factors affecting their distributions. Samples were collected from 965 sampling sites in streams across South Korea. We also measured 34 environmental variables grouped into six categories: geography, meteorology, land use, substrate composition, hydrology, and physicochemistry. A total of 83 taxa belonging to 10 families of Odonata were recorded in the dataset. Among them, eight species displayed high abundances and incidences. Self-organizing map (SOM) classified sampling sites into seven clusters (A⁻G) which could be divided into two distinct groups (A⁻C and D⁻G) according to the similarities of their odonate assemblages. Clusters A⁻C were characterized by members of the suborder Anisoptera, whereas clusters D⁻G were characterized by the suborder Zygoptera. Non-metric multidimensional scaling (NMDS) identified forest (%), altitude, and cobble (%) in substrata as the most influential environmental factors determining odonate assemblage compositions. Our results emphasize the importance of habitat heterogeneity by demonstrating its effect on odonate assemblages.

A Non-Parametric Approach for the Activation Detection of Block Design fMRI Simulated Data Using Self-Organizing Maps and Support Vector Machine

Functional magnetic resonance imaging (fMRI) is a popular method to probe the functional organization of the brain using hemodynamic responses. In this method, volume images of the entire brain are obtained with a very good spatial resolution and low temporal resolution. However, they always suffer from high dimensionality in the face of classification algorithms. In this work, we combine a support vector machine (SVM) with a self-organizing map (SOM) for having a feature-based classification by using SVM. Then, a linear kernel SVM is used for detecting the active areas. Here, we use SOM for feature extracting and labeling the datasets. SOM has two major advances: (i) it reduces dimension of data sets for having less computational complexity and (ii) it is useful for identifying brain regions with small onset differences in hemodynamic responses. Our non-parametric model is compared with parametric and non-parametric methods. We use simulated fMRI data sets and block design inputs in this paper and consider the contrast to noise ratio (CNR) value equal to 0.6 for simulated datasets. fMRI simulated dataset has contrast 1-4% in active areas. The accuracy of our proposed method is 93.63% and the error rate is 6.37%.

Self Organizing Map-Based Classification of Cathepsin k and S Inhibitors with Different Selectivity Profiles Using Different Structural Molecular Fingerprints: Design and Application for Discovery of Novel Hits

The main step in a successful drug discovery pipeline is the identification of small potent compounds that selectively bind to the target of interest with high affinity. However, there is still a shortage of efficient and accurate computational methods with powerful capability to study and hence predict compound selectivity properties. In this work, we propose an affordable machine learning method to perform compound selectivity classification and prediction. For this purpose, we have collected compounds with reported activity and built a selectivity database formed of 153 cathepsin K and S inhibitors that are considered of medicinal interest. This database has three compound sets, two K/S and S/K selective ones and one non-selective KS one. We have subjected this database to the selectivity classification tool ‘Emergent Self-Organizing Maps’ for exploring its capability to differentiate selective cathepsin inhibitors for one target over the other. The method exhibited good clustering performance for selective ligands with high accuracy (up to 100 %). Among the possibilites, BAPs and MACCS molecular structural fingerprints were used for such a classification. The results exhibited the ability of the method for structure-selectivity relationship interpretation and selectivity markers were identified for the design of further novel inhibitors with high activity and target selectivity.

Computational models on quantitative prediction of bioactivity of HIV-1 integrase 3' processing inhibitors

In this study, four computational quantitative structure-activity relationship (QSAR) models were built to predict the bioactivity of 3' processing (3'P) inhibitors of HIV-1 integrase. Some 453 inhibitors whose bioactivity values were detected by the radiolabelling method were collected. The molecular structures were represented with MOE descriptors. In total, 21 descriptors were selected for modelling. All inhibitors were divided into a training set and a test set with two methods: (1) by a Kohonen's self-organizing map (SOM); (2) by a random selection. For every training set and test set, a multilinear regression (MLR) analysis and a support vector machine (SVM) were used to establish models, respectively. For the training/test set divided by SOM, the correlation coefficients (r) were over 0.84, and for the training/test set split randomly, the r values were over 0.86. Some molecular properties such as hydrogen bond donor capacity, atomic partial charge properties, molecular refractivity, the number of aromatic bonds and molecular surface area, volume and shape properties played important roles for inhibiting 3' processing step of HIV-1 integrase.