Probabilistic Modeling and Machine Learning in Structural and Systems Biology

Completion of biological networks : the output kernel trees approach

author: Florence d'Alché, Université Evry Val d'Essonne

Description

Elucidating biological networks appears nowadays as one of the most important challenge in systems biology. Due to the availability of various sources of data, machine learning has to play a major role regarding this issue, given its large spectrum of tools ranging from generative models to concept learning methods. In this work the focus is narrowed on the completion of biological interactions networks for which some of the interactions between variables (usually genes or proteins) are already known.

Categories

Top: Computer Science: Machine Learning: Structured Output
Top: Computer Science: Machine Learning: Kernel Methods
Top: Computer Science: Bioinformatics

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Completion of biological networks the output kernel
0:22	Introduction
2:39	Outline
3:05	Outline - Supervised network inference
3:07	Supervised graph inference
4:04	A general solution based on considering a kernelized
5:21	A kernel on graph nodes
5:55	Outline - Output kernel trees
5:57	Output kernel trees
7:16	Back to the classic framework of supervised learning
7:51	Regression trees on multiple outputs
9:21	Multiple outputs regression trees
10:20	Prediction with multiple outputs regression trees
10:53	Output kernel trees (OK3) use the kernel trick in the output space pt 1
11:47	Output kernel trees (OK3) use the kernel trick in the output space pt 2
12:47	Algorithm devoted to the supervised network inference
14:20	Regression trees in output feature space pt 1
15:05	Regression trees in output feature space pt 2
16:09	Ensemble methods
19:04	Outline - Results
19:05	Biological networks
21:52	Comparison of different tree based methods
23:01	Comparison of different sets of features
23:42	Comparison with full kernel based methods
26:18	Robustness
27:47	Interpretability rules and clusters (an example with a protein-protein network)
30:45	Interpretability feature ranking
31:15	Conclusion
33:18	Interpretability rules and clusters (an example with a protein-protein network) (a)
35:15	Regression trees in output feature space pt 1 (a)
36:54	Comparison with full kernel based methods (a)

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.