User:Matthew Whiteside/Notebook/Malaria Microarray/2009/09/11

{| width="800"
 * style="background-color: #EEE"|[[Image:owwnotebook_icon.png|128px]] Project name
 * style="background-color: #F2F2F2" align="center"|  |Main project page
 * style="background-color: #F2F2F2" align="center"|  |Main project page


 * colspan="2"|
 * colspan="2"|

Mouse Malaria Microarray Datasets to Use in Meta-Analysis

 * Next step in the meta-analysis - is selection of the studies to use.
 * Selection will depend on the question being addressed (see task page).

Contrasts of Interest
To address the research question, i will look at two contrasts in gene expression: Brain gene expression will be compared. Here late-stage is onset of CM symptoms around day 6
 * 1) Late-stage CM infected mice vs. Late stage non-CM infected mice.
 * 2) Late-stage CM infected mice vs. uninfected baseline

The following acronyms will be used:
 * CM - CM infected mice
 * NCM - Non-CM infected mice
 * B - healthy baseline

Studies
Outlined below are the arrays that will be used to compute the contrasts (arrays classified as either CM, NCM or B) - UPDATE m2 and m3 were not carried forward, see Ref for explanation.

I have avoided custom cDNA arrays to keep things simple (its easier to annotate affy arrays). However in the future it may be necessary to include these too, in which case i would use the GEO array annotations.

Analysis Plans
With the two contrasts, i am planning to peform two different analyses


 * 1) CM vs NCM
 * 2) * Compute a PLD matrix for each study. Compute OR pathways for each study. Compare pathways.
 * 3) * I believe this approach is necessary because of the differences in NCM arrays - one study uses different malaria strains the other uses different mouse strains.
 * 4) CM vs B
 * 5) * Compute PLD matrix for each study. Combine PLD matrices into z-scores using meta-analysis.
 * 6) * This will provide a more statistical sound result and is possible because of the similar malaria infection models used in the three studies.