1. Gene-Disease Associations Based on network
报告人：李金金

2. Heterogeneous network
Contents
Background 1
Heterogeneous network 2
Methods 3

3. Background
Correctly identifying association of genes with diseases has long been a goal in biology.
Identifying association of genes with diseases has contributed to improving medical care and understanding of gene functions and interactions.
Clinical diseases are characterized by distinct phenotypes. To identity disease genes, the relationship between genes and phenotypes is involved.

4. Background
Problems
Pheno
-type
Gene
Disea
-se
Association

5. Construction heterogeous network
Gene network based on HPRD g3 g1 g2 g5 g6 g4 g7

6. Construction heterogeous network
Phenotype network using MinMiner p1 p2 p4 p5

7. Construction heterogeous network
Gene-Phenotype network based on OMIM p1 p2 p4 p5 g3 g1 g2 g5 g6 g4 g7

8. Construction heterogeous network

9. Methods
Katz
CATAPULT
Methods
CIPHER
GeneWalker
Prince
RWRH

10. Methods Methods Katz CATAPULT
is successfully applied for link prediction in social networks.
CATAPULT
is a supervised learning method.
Features are derived from hybrid walks through the heterogeneous network.

11. Katz g1 g2 g3 g4 g5 g6

12. Katz g1 g1 g2 g2 g3 g3 g3 g4 g5 g5 g6 g6 ……

13. Katz How to get the similarity matrix? Katz measure:，
Small values of k (k=3 or k=4) are known to yield competitive performance in the task of recommending similar nodes.

14. Katz on the heterogeneous network
Adjacency matrix of heterogeneous network: AG
gene-gene network B
the bipartite network genes and phenotypes
APHS
the similarity matrix of human diseases
APS
the similarity matrix of phenotypes of other species

15. Katz on the heterogeneous network
Katz similarity measure specialized to A:
K=3,the similarities between gene nodes and human disease nodes could be denoted by

16. the number of bootstraps
CATAPULT
How to train a biased SVM? T
the number of bootstraps
the sets of positive
the set of unlabeled gene-phenotype pairs n+
the number of examples in A
Step 1: Draw a bootstrap sample
of size n+ .
Step 2: Train a linear classifier θ using the positive training
examples A and U as negative examples.

17. CATAPULT
How to train a biased SVM?
Step 2: Training classifier

18. CATAPULT
How to train a biased SVM?
Step 3: For any
update:

19. Thank You !