[BioC] Affy & Probe packages : comparaison between the order of the probe in an affybatch and a probePackage.

[Laurent Buffat]

Hi Everybody,

 

I have a problem with the probe name (not the probe set name, but the probe
name, i.e. “probeSetName+Indice”, like 1431_at1) and the order of probe
between an affybatch and between the probe package information.

 

For example with the 133A chips and the probe set “1431_at”:

 

With the “hgu133aprobe” package, the order, for the 1431_at  is:

 

>
as.data.frame(hgu133aprobe[hgu133aprobe$Probe.Set.Name=="1431_at",c(2,3,4,5)
])

      x   y Probe.Set.Name Probe.Interrogation.Position

143 534 549        1431_at                         1013

144 653 237        1431_at                         1067

145 160 593        1431_at                         1073

146 324  61        1431_at                         1079

147 447 599        1431_at                         1115

148 327 395        1431_at                         1133

149 183 335        1431_at                         1169

150 652 615        1431_at                         1175

151 314 225        1431_at                         1187

152 437  57        1431_at                         1277

153 151 575        1431_at                         1361

154 113  39        1431_at                          941

155 677 607        1431_at                          947

156 120 695        1431_at                          953

157 248 585        1431_at                          959

158 157  93        1431_at                          965

 

If I read an affyBatch without normalisation and background correction, and
I look for the pm of “1431_at” for the first and second experiment:

 

> pm(anAffyBatch,"1431_at")[,c(1,2)]

               1      2

1431_at1  2692.0 2837.0

1431_at2   252.8  340.8

1431_at3   105.3  106.0

1431_at4   112.0  133.0

1431_at5   243.0  226.3

1431_at6   952.3  790.0

1431_at7   707.5  806.0

1431_at8   520.8  687.3

1431_at9  2297.5 2797.8

1431_at10 1109.8 1218.0

1431_at11  594.3  757.8

1431_at12  560.8  563.0

1431_at13  950.0 1131.5

1431_at14  899.5  768.0

1431_at15  347.0  377.8

1431_at16 1628.0 2209.5

 

And now, if I read in a text editor (like emacs) the CEL file corresponding
for the first experiment, and I check for the signal at (x,y) = (534 549),
corresponding to the “first” probe in the hgu133aprobe , I obtained the
line:

 

534      549      952.3   169.3   16 113            39       2692.0 195.2
16      

 

And the signal “952.3” is not the signal of the “first” probe 1431_at1 in
the affybatch what is “2692.0”, but correspond to the 6th in the affybatch :
1431_at6.

 

If I search for the signal “2692.0” (the first PM in the affyBatch)  in the
CEL file, I have the line :

 

113      39       2692.0 195.2   16

 

And the coordinates (113,39) correspond  of the 12th “row” in the
hgu133aprobe.

 

If you look at the ProbeInterogationPostition, in the hgu133aprobe, you have
a “gap” between the 11th and 12th and if you order the information according
to the ProbeInterogationPostion, when you obtained:

 

> order(hgu133aprobe[hgu133aprobe$Probe.Set.Name=="1431_at",5])

 [1] 12 13 14 15 16  1  2  3  4  5  6  7  8  9 10 11

 

And now, this order corresponds to the order in the affybatch, for this
particular probeSet.

 

So I have some questions:

 

1/ It’s seems that the order in an affybatch is based on the
Probe.Intergogation.Position ?

Is it true? Will it be always like that?

 

2/ I have no idea about the order in the hgu133aprobe. How is-it calculated?

 

3/ Is it possible to have a “harmonisation” of the two orders?

 

4/ How is it possible to simple merge an affybatch (the pm) and an
hgu133aprobe

(actually, I ordered the hgu133aprobe according to the
Probe.Interrogation.Position and cbind the two data.frame, but it’s work
only if the order in an affybatch is really based on this).

 

Thanks for your helps,

 

Bests.

 

Laurent Buffat

 

 

 

 

Laurent Buffat MD, PhD

CSO It-Omics

Parc Eurasanté

885, avenue Eugène Avinée

59120 Loos

Tél : 33 (0)3 20 16 40 56

 

 


	[[alternative HTML version deleted]]