Microbial secondary metabolites are a potent source of antibiotics and other pharmaceuticals. Genome mining their biosynthetic gene clusters has become key method to accelerate identification characterization. In 2011, we developed antiSMASH, web-based analysis platform that automates this process. Here, present the highly improved antiSMASH 2.0 release, available at http://antismash.secondarymetabolites.org/. For new version, was entirely re-designed using plug-and-play concept allows easy...

Abstract. In the framework of a changing climate, it is useful to devise methods capable effectively assessing and monitoring landscape air–sea CO2 fluxes. this study, we developed an integrated machine learning tool objectively classify track marine carbon biomes under seasonally interannually environmental conditions. The was applied monthly output global ocean biogeochemistry model at 0.25° resolution run atmospheric forcing for period 1958–2018. Carbon are defined as regions having...

Lanthipeptides are a class of ribosomally synthesised and post-translationally modified peptide (RiPP) natural products from the bacterial secondary metabolism. Their name is derived characteristic lanthionine or methyl-lanthionine residues contained in processed peptide. that possess an antibacterial activity called lantibiotics. Whereas multiple tools exist to identify lanthipeptide gene clusters genomic data, no programs available predict post-translational modifications lanthipeptides,...

Our research focuses on the detection of ocean carbon uptake regimes that are critical in context comprehending climate change. One observation among geoscientific data Earth System Sciences is datasets often contain local and distinct statistical distributions posing a major challenge applying clustering algorithms for analysis. The use global parameters many inadequate to capture such distributions. In this study, we propose novel tool detect visualize oceanic clusters. We implement...

Our research focuses on detecting and tracking ocean carbon regimes, which are crucial indicators for understanding the impacts of climate change uptake. Geoscientific datasets in Earth System Sciences often contain local distinct statistical distributions at a regional scale. This poses significant challenge applying conventional clustering algorithms data analysis. Based observed limitations prominent methods, our study, we propose framework that enhances well-established unsupervised...

Abstract. In the framework of a changing climate, it is useful to devise methods capable effectively assessing and monitoring landscape air-sea CO2 fluxes. this study, we developed an integrated machine learning tool objectively classify track marine carbon biomes under seasonally interannually environmental conditions. The was applied monthly output global ocean biogeochemistry model at 0.25° resolution run atmospheric forcing for period 1958–2018. Carbon are defined as regions having...

Abstract We discuss topological aspects of cluster analysis and show that inferring the structure a dataset before clustering it can considerably enhance detection: we embedding vectors representing inherent instead observed feature themselves is highly beneficial. To demonstrate, combine manifold learning method UMAP for with density-based DBSCAN. Synthetic real data results this both simplifies improves in diverse set low- high-dimensional problems including clusters varying density and/or...

The increasing digital traces of fishing fleets nowadays available allow for automatized observation the oceans, a vulnerable space which could hardly be monitored or governed previously. Data streams from satellite base communication systems are being used variety applications such as collision avoidance, route optimization, and monitoring illegal activities.

Correlation clustering detects complex and intricate relationships in high-dimensional data by identifying groups of points, each characterized differents correlation among a (sub)set features. Current methods generally limit themselves to linear correlations only. In this paper, we introduce method for detecting global non-linear correlated clusters focusing on quadratic relations. We novel Hough transform the detection hyperparaboloids apply it hyperparaboloid arbitrary spaces. Non-linear...

Abstract Unsupervised learning methods are well established in the area of anomaly detection and achieve state art performances on outlier datasets. Outliers play a significant role, since they bear potential to distort predictions machine algorithm given dataset. Especially among PCA-based methods, outliers have an additional destructive regarding result: may not only orientation translation principal components, also make it more complicated detect outliers. To address this problem, we...

In 2017 Day et al. introduced the notion of locality as a structural complexity-measure for patterns in field pattern matching established by Angluin 1980. 2019 Casel showed that determining an arbitrary is NP-complete. Inspired hierarchical clustering, we extend to coloured graphs, i.e., given graph determine enumeration colours such colouring stepwise according leads few clusters possible. Next first theoretical results on classes, propose priority search algorithm compute $k$-locality...

Analyzing flyways of birds is one approach ornithologists pursue e.g. to be able detect potential risks during the animal's migration. But this analysis not trivial and functionalities existing supporting tools are neither perfect nor all-encompassing. In paper, we introduce our new FATBIRD Tool, which only visualizes or arbitrary trajectories, but also helps researchers in several aspects analysis. Similarities between all trajectories individual calculated via Dynamic Time Warping...

LUCK allows to use any distance-based clustering algorithm find linear correlated data. For that a novel distance function is introduced, which takes the distribution of kNN points into account and corresponds probability two being part same correlation. In this work in progress we tested measure with DBSCAN k-Means comparing it well-known correlation algorithms ORCLUS, 4C, COPAC, LMCLUS, CASH, receiving good results for difficult synthetic data sets containing crossing or non-continuous...

In this work we propose SRE, the first internal evaluation measure for arbitrary oriented subspace clustering results. For purpose present a new perspective on task: goal formalize is to compute which represents original dataset by minimizing reconstruction loss from obtained subspaces, while at same time dimensionality as well number of clusters. A fundamental feature our approach that it model-agnostic, i.e., independent characteristics any specific method. It scale invariant and...