0.25
0.5
0.75
1.25
1.5
1.75
2
Evidence Integration in Bioinformatics
Published on Feb 25, 20075132 Views
Biologists frequently use databases; for example, when a biologist encounters some unfamiliar proteins, s/he will use databases to get a preliminary idea of what is known about them. The databases can
Related categories
Chapter list
Evidence Integration in Bioinformatics00:00
A little molecular biology00:13
Problems01:26
Evidence of related function01:45
Evidence of protein-protein interaction02:40
Combining using machine learning04:19
Overfitting and inductive bias06:03
Supervised Learning with Bayes Nets07:07
Bayesian Networks 07:19
Bayes Net - Example08:04
Naïve Bayes10:22
Hierarchical Naïve Bayes13:49
Supervised learning with SVMs (kernel fusion)15:16
Support Vector Machines for Classification15:26
Support Vector Machine Training17:34
Kernel fusion18:43
Evaluation - ROC Curve21:34
Evaluation - ROC curve23:01
Results – membrane protein prediction23:38
Supervised learning with boosting (RankBoost)24:45
RankBoost25:00
RankBoost behavior26:30
RankBoost28:17
Unsupervised Learning with Bayes Nets29:05
Regulation of Expression30:00
Transcriptional Modules31:19
Bayes Net for Transcription Modules33:06
Unsupervised Evidence Integration37:54
Problem37:57
Examples38:20
More generally41:36
Notation42:22
Isn’t this just clustering?43:15
Related Theoretical Work [MV03] – Problem45:17
Related Theoretical Work [MV03] – Results47:22
In our problem(s)...49:24
Conditional independence50:30
Our Approach52:44
Notes58:56
Evaluation: Yeast protein-protein data59:14
Evaluation: other algorithms01:01:10
Evaluation01:01:36
Results: Protein-protein data01:01:57
Results: Protein-protein data01:02:46
Evaluation: Artificial data01:03:49
Results: Artificial source01:04:07
Results: Artificial source01:04:09
Paper and Software01:04:12