Data integration is a hot topic in bioinformatics, but the term means different things to different people. What do we think it means? Talk given at CSIRO Bioinformatics & Biostatistics group meeting, November 21 2012.
12. Quote from integIRTy paper
These methods can be roughly grouped into four categories:
stepwise, regression-based, correlation-based and
latent variable models
integIRTy: a method to identify genes altered in cancer by accounting for
multiple mechanisms of regulation using item response theory
Bioinformatics, Vol. 28, No. 22. (15 November 2012), pp. 2861-2869
Data integration 12 of 21
13. Regression: SIM
Integrated analysis of DNA copy number and gene expression microarray data using gene sets
BMC Bioinformatics 2009, 10:203
Data integration 13 of 21
16. Basics that are never explained 1/2
Integration across groups or description of samples?
Data integration 16 of 21
17. Basics that are never explained 2/2
Genes x Samples
Data integration 17 of 21
18. Conclusions 1/3
We’re not the first people doing this...
...but it’s becoming a “hot topic”
Data integration 18 of 21
19. Conclusions 2/3
Room for improvement in software, much of which is:
• Poorly-written
• Poorly-documented
• Difficult to implement
Data integration 19 of 21
21. CSIRO Mathematics, Informatics and Statistics
Neil Saunders
t
+61 2 9325 3144
e Neil.Saunders@csiro.au
w Mathematics, Informatics and Statistics web
MATHEMATICS, INFORMATICS AND STATISTICS
www.csiro.au