Quick Guide to Using Affymetrix Microarray Suite (MAS 5.0)

There is a more detailed explanation of each of these steps in the black BCF binder in Keating 111D.  The paper "GeneChip Expression Analysis - Data Analysis Fundamentals" contains much useful information not found in the Affymetrix manuals.  Refer to this paper when evaluating your analysis reports and for instructions on identifying robust Increases/Decreases and Fold Change calculations. 

1)      Copy the data files (.DAT, .CEL, .EXP, .CHP) into your folder on Brahms, or to C:\GeneChip\TEMPDATA. These are the raw data files and Absolute analyses from the GATC microarray core facility.

2)      Start->Programs->Affymetrix->MicroArraySuite  and verify these settings

In Tools->Defaults:

File Locations: Modify Experiment Data line to refer to the folder that contains your Affy data files and the Library location to refer to the correct Chip Library

Analysis Settings: Enable 'Prompt for output filename' and 'Display settings when analyzing data'

In Tools->Analysis Settings->Expression:

Select the Probe Array Type

Default Scaling is ‘All Probe Sets’, Target 500

Default Normalization is ‘User Defined’ value 1

Default Baseline is OFF (No Comparison File for absolute analyses)

3)      Open each .DAT file and check the gridding. It is always a good idea to look at the raw data!

4)      Generate an absolute analysis for each .CEL file (or use the existing .CHP files). The same "Target Intensity" value must be used for Scaling all Absolute Analyses so that comparisons will be meaningful!!

5)      Generate a report from each .CHP file and check quality and controls. See Chp 1 of Data Analysis Fundamentals, Guidelines for Assessing Sample and Array Quality.

6)      Generate Comparison analyses from the Batch Analysis Window. Drag the .CHP files into the Batch Analysis Window, set the output .CHP file names and Baseline .CHP files for each comparison.  When the comparisons finish, select a baseline and comparison pair and open them together to see the Change Calls and Signal Log Ratios.

7)      Filter out all A->A calls by Sorting and Hiding. Press Sort button, select Detection Ascending for baseline and comparison analyses, Hide Selected.

8)      Identify Robust Increases and Decreases. The steps are listed in the Data Analysis Fundamentals guide.  After filtering the data, you can Export it to Excel to analyze it further.

9)      Scatter Plot for Fold Change Estimate. Sort the pivot table and hide everything other than P->P, the press the "Scatter graph" button. This graph displays only the simple Signal Ratios.  For a more accurate Fold Change calculation, you will need to paste the data into Excel and use the formulas shown on pages 11-12 of the Data Analysis Fundamentals guide. You can right-click->Copy the image into the clipboard and paste it into PowerPoint.

10)   If you are looking at a Time Series experiment, create a Bar or Line graph and arrange the time points across the X-axis.

11)   Access NetAffx to see additional information about probe sets.

12)   When you have finished, save your results on a CD.  The C:\TEMPDATA folder is emptied nightly and your Brahms folder cannot hold more than a few analysis files.