The fulltext of publications might not be freely accessible but require subscription. Please request reprints at

Publications in peer reviewed journals

2 Publications found
  • scikit-hubness: Hubness Reduction and Approximate Neighbor Search

    Feldbauer R, Rattei T, Flexer A
    2020 - The Journal of Open Source Software, 5: 1957


    scikit-hubness is a Python package for efficient nearest neighbor search in high-dimensional spaces. Hubness is an aspect of the curse of dimensionality in nearest neighbor graphs. Specifically, it describes the increasing occurrence of hubs and antihubs with growing data dimensionality: Hubs are objects, that appear unexpectedly often among the nearest neighbors of others objects, while antihubs are never retrieved as neighbors. As a consequence, hubs may propagate their information (for example, class labels) too widely within the neighbor graph, while information from antihubs is depleted. These semantically distorted graphs can reduce learning performance in various tasks, such as classification, clustering, or visualization. Hubness is known to affect a variety of applied learning systems, or improper transport mode detection.

    Currently, there is a lack of fully-featured, up-to-date, user-friendly software dealing with hubness. Available packages miss critical features and have not been updated in years, or are not particularly user-friendly. In this paper we describe scikit-hubness, which provides powerful, readily available, and easy-to-use hubness-related methods.

  • DeepNOG: Fast and accurate protein orthologous group assignment.

    Feldbauer R, Gosch L, Lüftinger L, Hyden P, Flexer A, Rattei T
    2020 - Bioinformatics, in press


    Protein orthologous group databases are powerful tools for evolutionary analysis, functional annotation, or metabolic pathway modeling across lineages. Sequences are typically assigned to orthologous groups with alignment-based methods, such as profile hidden Markov models, which has become a computational bottleneck.
    We present DeepNOG, an extremely fast and accurate, alignment-free orthology assignment method based on deep convolutional networks. We compare DeepNOG against state-of-the-art alignment-based (HMMER, DIAMOND) and alignment-free methods (DeepFam) on two orthology databases (COG, eggNOG 5). DeepNOG can be scaled to large orthology databases like eggNOG, for which it outperforms DeepFam in terms of precision and recall by large margins. While alignment-based methods still provide the most accurate assignments among the investigated methods, computing time of DeepNOG is an order of magnitude lower on CPUs. Optional GPU usage further increases throughput massively. A command-line tool enables rapid adoption by users.
    Source code and packages are freely available at Install the platform-independent Python program with $pip install deepnog.
    Supplementary material is available at Bioinformatics online.

Book chapters and other publications

No matching database entries were found.