GraphKKE: graph Kernel Koopman embedding for human microbiome analysis

  title={GraphKKE: graph Kernel Koopman embedding for human microbiome analysis},
  author={Kateryna Melnyk and Stefan Klus and Gr{\'e}goire Montavon and Tim O. F. Conrad},
  journal={Applied Network Science},
More and more diseases have been found to be strongly correlated with disturbances in the microbiome constitution, e.g., obesity, diabetes, or some cancer types. Thanks to modern high-throughput omics technologies, it becomes possible to directly analyze human microbiome and its influence on the health status. Microbial communities are monitored over long periods of time and the associations between their members are explored. These relationships can be described by a time-evolving graph. In… 

Understanding microbiome dynamics via interpretable graph representation learning

Through the experiments, this work shows that it can extract graph features such as clusters of nodes or edges that have the highest impact on the model to learn the low-dimensional representation of the time-evolving graph while maintaining the dynamics occurring in the high-dimensional space.

Multiple Kernel Representation Learning on Networks

A weighted matrix factorization model that encodes random walk-based information about nodes of the network and extends the approach with a multiple kernel learning formulation that provides the ability to utilize kernel functions without realizing the exact proximity matrix.

Representation Learning for Networks in Biology and Medicine: Advancements, Challenges, and Opportunities

This review synthesizes a spectrum of algorithmic approaches that, at their core, leverage topological features to embed networks into compact vector spaces and provides a taxonomy of biomedical areas that are likely to benefit most from algorithmic innovation.



A kernel-based approach to molecular conformation analysis

This work shows that many of the prominent methods like Markov state models, extended dynamic mode decomposition (EDMD), and time-lagged independent component analysis (TICA) can be regarded as special cases of this approach and that new efficient algorithms can be constructed based on this derivation.

The Role of the Gut Microbiome in Colorectal Cancer Development and Therapy Response

A personalized modulation of the pattern of gut microbiome by diet may be a promising approach to prevent the development and progression of CRC and to improve the efficacy of antitumoral therapy.

Microbiome‐immune‐metabolic axis in the epidemic of childhood obesity: Evidence and opportunities

  • H. KincaidR. NagpalH. Yadav
  • Medicine, Biology
    Obesity reviews : an official journal of the International Association for the Study of Obesity
  • 2019
How factors from as early as gestation appear to contribute in obesity, such as maternal health, diet, antibiotic use by mother and/or child, and birth and feeding methods are discussed and gaps in knowledge are described.

Metastability: A Potential-Theoretic Approach

Metastability is an ubiquitous phenomenon of the dynamical behaviour of complex systems. In this talk, I describe recent attempts towards a model-independent approach to metastability in the context

Kernel methods for detecting coherent structures in dynamical data.

This work shows that kernel canonical correlation analysis (CCA) can be interpreted in terms of kernel transfer operators and that it can be obtained by optimizing the variational approach for Markov processes score, and demonstrates the efficiency of this approach with several examples.

Modelling microbiome recovery after antibiotics using a stability landscape framework

A simple quantitative model based on the stability landscape concept is outlined and support for a long-term transition to an alternative microbiome state after courses of certain antibiotics in both the gut and oral microbiomes is found.

The gut microbiome: Relationships with disease and opportunities for therapy

This review provides an overview of the influence of the gut microbiome on host health with a focus on immunomodulation and discusses strategies for manipulating the gut microbiome for the management

MetaNN: accurate classification of host phenotypes from metagenomic data using neural networks

MetaNN is proposed, a neural network framework which utilizes a new data augmentation technique to mitigate the effects of data over-fitting and outperforms existing state-of-the-art models in terms of classification accuracy for both synthetic and real metagenomic data.

Microbiome functioning depends on individual and interactive effects of the environment and community structure

Surprisingly, the effects of taxonomic diversity and normalized oxidase abundance in the model predicting CO2 production were attributable to site-level differences in bacterial communities unrelated to the present-day environment, suggesting that colonization history rather than habitat-based filtering indirectly influenced ecosystem functioning.