Analysis of Agilent Slides

The analysis of DNA microarray data can be done using various methods. In this analysis, standard routines included in the LimmaR package are used. Although most methods require input parameters, the Genome2D DNA microarray webserver uses predefined settings. This means that the user only needs to have basic statistical knowledge. The data will be normalized using Lowess normalization, followed by an in-between-slides quantile normalization. Subsequently, for each array the quality will be examined to add a weight factor. Graphs and tables will be generated automatically.

Headers of the data table are important and should exactly be:

R="F635 Mean",G="F532 Mean",Rb="B635 Median",Gb="B532
R="F650 Mean",G="F550 Mean",Rb="B650 Median",Gb="B550
R="Raw intensity (tr.mean) {B}", Rb="Background (tr.mean) {B}", G="Raw intensity (tr.mean) {A}", Gb="Background (tr.mean) {A}",annotation="Block","Row","Column","ID","Name"
R="rMedianSignal", Rb="rBGMedianSignal", G="gMedianSignal", Gb="gBGMedianSignal", annotation="Block","Row","Column","ID","Name"

Step 1: Selecting files for the analysis

A) Pack all: raw data files ( described in Targets.txt) into one ZIP file and select the file type below
           GenePix (standard headers 635/532) ( NOTE: Split the 8x15k to 1x15k arrays here )
           GenePix (headers 650/550) typically the image is scanned on an Agilent scanner ( NOTE: Split the 8x15k to 1x15k arrays here )
B) Make a tab-delimited file called "Targets.txt" which contains your hybridisation scheme and the file names of the raw data
          ==> example file
Targets file:
C) Make a tab-delimited file "SpotTypes.txt" that contains genes you want to highlight in your analysis
          ==> example file
SpotTypes file:
D) Make / edit a tab-delimited file "Contrasts.txt". Here you describe what to compare. E.g., if the common reference is called MyControl, the contrasts lines can be: sampleB, sampleC, sampleD to compare these samples with the common reference MyControl. To compare sampleB with sampleD, just type in one line; sampleB-sampleD. Technical and/or biological replicates can be named similar and this will be handled as standard replicates in the analysis. Alternatively, to each replicate an unique name can be assigned, but then you must indicate the replicates in the Contrasts.txt file: sampleD-(sampleB+sampleC)/2 means: compare D with biological replicates B and C.
          ==> example file
Contrasts file:

Step 2: Common Reference and Experiment name

The name of the Common Reference (as mentioned in Targets.txt): Reference name:
The name of the experiment will be attached to all restuls of this analysis. Experiment name:

Step 3: Type of analysis

Paired samples analysis using CyberT: e.g., wt versus mutant
Common reference method of LimmaR (this is currently the only supported method)
Factorial design: e.g., wt versus mutant versus oxygen versus no oxygen

Table 1
Links to files generated can be downloaded by following the links.