Why are there more spike-ins?

Prepared by Zhijin Wu

We add two more probesets than what Affymetrix does into the spike-in list, "33818_at" and "546_at". Here's why:

1. We claim "33818_at" should be in Latin Square as Transcript #12. What AFFY gives in the position of Transcript #12 is "407_at", which actually has the same concentration pattern as Transcript #1, "37777_at". No spikein gene given by AFFY has the pattern in the 12th column of the Latin Square.
However, "33818_at" has the pattern missing from the Latin Square, agreed by three methods of calculating expression (RMA, MAS 5.0, dChip). Wolfinger et al (CAMDA 2002) identified this as well.

2. We claim "546_at" should be considered with same concentration as "36202_at", because it's designed against the same target, Unigene ID Hs. 75209 It has the pattern same as "36202_at", as shown by three methods. Wolfinger et al identified this as well. (page 2,3,4 in file extraspk.pdf)

3. Wolfinger et al. claim "1598_g_at" and "37658_at" have same pattern as "1597_at" . This is not seen in rma, somewhat in mas, clear in lw expression. (page 5,6,7 in file extraspk.pdf) 4. WOlfinger claims "1032_at" has same pattern as "684_at" This is not seen in rma or mas, but clear in lw expression. (page 8,9,10 in file extraspk.pdf)


Back to Home Page [send me mail]
You are visitor number