en
0.25
0.5
0.75
1.25
1.5
1.75
2
Accuracy test for genome wide selection of bio-markers
Published on Jan 23, 20123449 Views
Biochemistry of the living cell involves multitude of very complex nonlinear interactions between proteins coded for by DNA. Thus the in-silico searches for disease related bio-markers have to consid
Related categories
Chapter list
Accuracy test for genome wide selection of bio-markers00:00
Overview00:23
What is the problem?01:45
Background and Motivation02:48
Size of Epistasis Search Space03:01
Voyager 204:29
Does It Work? Some Timing Data04:53
Biomarker Selection Main Challenge: low n high p07:03
Binomial Margin09:14
Case-Control Study09:24
Binomial Margin - a novel Hypothesis Test10:54
Computation of Binomial Margin (a simple case of lower and upper bound)14:57
Computation of Binomial Margin (a closed form case)15:18
Another bound15:22
A more involved solution15:24
Comparison of Bounds for Binomial Margin15:42
Relation to Fisher Exact test: we are more conservative!16:26
Quantification of interaction effect16:59
Strong Epistasis indicates need for exhaustive search19:40
Replication of interactions in 5 Celiac cohorts21:20
Statistical Validation: replication in Celiac disease22:55
Some timing results for Celiac UK1 data23:31
Validation: Type 2 Diabetes and Celiac disease (UK1)24:08
Replicability of epistasis for different algorithms24:42
Replication for accuracy test with different epistasis filtering criteria25:34
Models of Interaction - 125:55
Models of Interaction - 226:53
Conclusions27:10
Acknowledgements28:16