Combined different data format and array types

[ccye...@gmail.com]

Dear Dr. Li:

I have downloaded a series of data from GEO sets:

Two sets: U133A, but only text data available
Two sets: U133A, CEL data available
Other sets: U133 plus 2.0, CEL data available

I have combined text format data from the 4 sets using U133A platform
(the so call "series_matrix data"), and used "get external data" read
into dChip and get normalized data. Then used 'export expression value
into text format

For U133 plus 2.0, I used dChip "open group" and read them all and get
normalized data. Then used 'export expression value" to get expression
value of U133A probes set

Then I combined these two sets of data into one, and use "get
external" function again to read into dChip for analysis.

Questions:
(1) Do you think what I did is appropriate?
(2) I used separator to treat U133A group and U133 2.0 group as
different batch to do ANOVA and clustering. Is that necessary? Or not
eough? Or should I just "normalized" all data and treat them the same?
(3) I used gene filter function to select around 12000 genes, and
export these genes into another set of data, and then read them into
dChip using "get external" again. Did I need to normalize these data
again in dChip?