printlogo
http://www.ethz.ch/index_EN
Machine Learning Laboratory
 
print
  

Computational Biology

Awards

Kay H. Brodersen received a Trainee Abstract Award at HBM 2012 for his work on 'Model-Based Clustering Using Generative Embedding.' Kay will be giving a talk on his results in Beijing on 12 June.

François Cellier received the McLeod Founder's Award of the Society for Modeling and Simulation International.

Spotlight

spotlight



Computational models of brain connectivity, coupled with machine learning algorithms, make it possible to infer neuronal disease mechanisms from non-invasive functional magnetic resonance imaging (fMRI) data in humans. This illustration shows how dynamic systems models can be used for reducing complex (high-dimensional) brain activity data to a simple (low-dimensional) and mechanistically interpretable representation (Brodersen et al., PLoS Comput. Biol. 2011). Please also see the summary on ETH Life.

Introduction

With the advent of high throughput methods for genomics and proteomics, automated tools for processing and analysis of the large amounts of data is highly necessary. Our efforts are focused on using machine learning and statistical techniques, to solve complex problems arising in biological and medical context. We emphasize on applications of kernel-based analysis, network inference and probabilistic modeling. The research is inspired by real-world problems and data, that arise from our established collaborations with biologists and medical scientists. In collaboration with other groups, we are further interested in using the predictions of our machine learning algorithms to guide biological experiments.  

Projects

Identifying protein complexes

Proteins are essential parts of organisms participating in every process within cells.  Vital cellular functions, such as DNA replication and mRNA translation, can only take place through the coordinated action of proteins that are assembled into a number of multi-protein complexes of varying composition and structure. Therefore, understanding how proteins interact and form functional modules is of central importance in biological research. Various large-scale efforts have thus attempted to probe the space of protein-protein interactions (PPI) in several organisms. Commonly used techniques are yeast two-hybrid (Y2H) and affinity purification coupled with mass spectrometry (AP-MS).

Our approach tries to derive a sensible error model for AP-MS data coupled with a clustering algorithm, in order to derive, in an unsupervised fashion, biologically-meaningful protein complexes.

   
xf_project
   
Contact: Xenofon Floros
   

Experiment design in system biology

Systems biology studies biological processes as dynamic, integrated networks of interacting molecules. These are complex systems whose mechanisms can, in principle, be mathematically modelled. Obtaining a reliable mathematical representation of biological systems is important for analysis, simulation and control. This task is challenging because of several levels of uncertainty both in reaction rates and in pathway connectivity. Machine learning provides modelling and computational approaches which enable the quantitative analysis of the observations and the design of maximally informative experiments.    
agb_project_1
   
The combination of model selection/parameter estimation and optimal experimental design close the loop between modelling and computation. New biological hypotheses are iteratively generated and tested.    
     
agb_project_2
   
Optimal experimental design provides the mathematical framework required for maximally informative experiments. It selects the most appropriate critical setting, together with the perturbation that permit to discriminate between alternative hypotheses.    
Contact: Alberto Giovanni Busetto
   
Collaborators:
* LiverX
* YeastX
* Competence Center for Systems Physiology and Metabolic Diseases (CC-SPMD)
 

Wichtiger Hinweis:
Diese Website wird in älteren Versionen von Netscape ohne graphische Elemente dargestellt. Die Funktionalität der Website ist aber trotzdem gewährleistet. Wenn Sie diese Website regelmässig benutzen, empfehlen wir Ihnen, auf Ihrem Computer einen aktuellen Browser zu installieren. Weitere Informationen finden Sie auf
folgender Seite.

Important Note:
The content in this site is accessible to any browser or Internet device, however, some graphics will display correctly only in the newer versions of Netscape. To get the most out of our site we suggest you upgrade to a newer browser.
More information

© 2012 ETH Zurich | Imprint | Disclaimer | 16 November 2011
top