[BioC] Loop design - biological, technical replication and contrasts
Maciej Jończyk
mjonczyk at biol.uw.edu.pl
Thu Feb 25 19:25:52 CET 2010
Dear list Members.
I'd like to ask for advice.
I have 2x2 factorial experiment conducted in a loop-design (two-colour
data).
There are two maize lines (dh7, dl3) and two temperature treatments
(cold=c and control=k).
Following hybridizations were conducted : dh_c vs dh_k; dh_k vs dl_k;
dl_k vs dl_c; dl_c vs dh_c (forming a square on a diagram)
and two (diagonal on a diagram), namely: dl_k vs dh_c and dh_k vs dl_c.
I have four biological replications of this design, including dye-swap
(i.e. two hyb. cy3-cy5 and two cy5-cy3).
In each biological replication I have also three technical replications
of each RNA source
(i.e. in dh_c vs dh_k; dl_c vs dh_c and dl_k vs dh_c, sample dh_c is
from the same RNA pool).
I don't know how analyse it in limma. I want to get following contrasts:
dh_c vs dh_k; dh_k vs dl_k; dl_k vs dl_c; dl_c vs dh_c and include
effect of temperature,
maize line and interaction.
My targets frame is:
SlideNumber FileName Cy3 Cy5
1 93 c_093_DH_K_vs_DH_CHex.gpr dh7_k_1 dh7_c_1
2 104 c_104_DH_CH_vs_DH_Kex.gpr dh7_c_2 dh7_k_2
3 116 c_116_DHK_vs_DHCHex.gpr dh7_k_3 dh7_c_3
4 16 c_016_DH_C_vs_DH_Kex.gpr dh7_c_4 dh7_k_4
5 94 c_094_DH_K_vs_DL_Kex.gpr dh7_k_1 dl3_k_1
6 105 c_105_DL_K_vs_DH_Kex.gpr dl3_k_2 dh7_k_2
7 117 c_117_DHK_vs_DLKex.gpr dh7_k_3 dl3_k_3
8 139 c_139_DL_K_vs_DH_Kex.gpr dl3_k_4 dh7_k_4
9 92 c_092_DL_CH_vs_DL_Kex.gpr dl3_c_1 dl3_k_1
10 106 c_106_DL_K_vs_DL_CHex.gpr dl3_k_2 dl3_c_2
11 118 c_118_DLCH_vs_DLKex.gpr dl3_c_3 dl3_k_3
12 23 c_023_DL_K_vs_DL_Cex.gpr dl3_k_4 dl3_c_4
13 95 c_095_DL_CH_vs_DH_CHex.gpr dl3_c_1 dh7_c_1
14 107 c_107_DH_CH_vs_DL_CHex.gpr dh7_c_2 dl3_c_2
15 119 c_119_DLCH_vs_DHCHex.gpr dl3_c_3 dh7_c_3
16 136 c_136_DH_C_vs_DL_Cex.gpr dh7_c_4 dl3_c_4
17 101 c_101_DL_K_vs_DH_CHex.gpr dl3_k_1 dh7_c_1
18 103 c_103_DH_CH_vs_DL_Kex.gpr dh7_c_2 dl3_k_2
19 121 c_121_DLK_vs_DHCHex.gpr dl3_k_3 dh7_c_3
20 15 c_015_DH_C_vs_DL_Kex.gpr dh7_c_4 dl3_k_4
21 100 c_100_DH_K_vs_DL_CHex.gpr dh7_k_1 dl3_c_1
22 102 c_102_DL_CH_vs_DH_Kex.gpr dl3_c_2 dh7_k_2
23 120 c_120_DHK_vs_DLCHex.gpr dh7_k_3 dl3_c_3
24 140 c_140_DL_C_vs_DH_Kex.gpr dl3_c_4 dh7_k_4
In "Cy3" and "Cy5" columns last number denotes biological replication.
I've read in limma documentation, that this desing can be analysed using
single channel analysis.
So I created following contrast matrix:
FileName dye dhc dhk dlc
c_093_DH_K_vs_DH_CHex.gpr 1 -1 1 0
c_104_DH_CH_vs_DH_Kex.gpr 1 1 -1 0
c_116_DHK_vs_DHCHex.gpr 1 -1 1 0
c_016_DH_C_vs_DH_Kex.gpr 1 1 -1 0
c_094_DH_K_vs_DL_Kex.gpr 1 0 1 0
c_105_DL_K_vs_DH_Kex.gpr 1 0 -1 0
c_117_DHK_vs_DLKex.gpr 1 0 1 0
c_139_DL_K_vs_DH_Kex.gpr 1 0 -1 0
c_092_DL_CH_vs_DL_Kex.gpr 1 0 0 1
c_106_DL_K_vs_DL_CHex.gpr 1 0 0 -1
c_118_DLCH_vs_DLKex.gpr 1 0 0 1
c_023_DL_K_vs_DL_Cex.gpr 1 0 0 -1
c_095_DL_CH_vs_DH_CHex.gpr 1 -1 0 1
c_107_DH_CH_vs_DL_CHex.gpr 1 1 0 -1
c_119_DLCH_vs_DHCHex.gpr 1 -1 0 1
c_136_DH_C_vs_DL_Cex.gpr 1 1 0 -1
c_101_DL_K_vs_DH_CHex.gpr 1 -1 0 0
c_103_DH_CH_vs_DL_Kex.gpr 1 1 0 0
c_121_DLK_vs_DHCHex.gpr 1 -1 0 0
c_015_DH_C_vs_DL_Kex.gpr 1 1 0 0
c_100_DH_K_vs_DL_CHex.gpr 1 0 1 -1
c_102_DL_CH_vs_DH_Kex.gpr 1 0 -1 1
c_120_DHK_vs_DLCHex.gpr 1 0 1 -1
c_140_DL_C_vs_DH_Kex.gpr 1 0 -1 1
As a reference sample I choosen "dlk", I've included also dye effect.
Now, I don't know how can I include information about technical
replication.
I assume that if I don't do it, it will be treated as biological
replication.
I read about command "duplicateCorrelation" is it the right one to deal
with these problem?
I also have problem in contrasts definition.
Is it simply dhc-dhk, dhk-dlk, dlk-dlc, dlc-dhc
and for main effect (maize line): (dhc+dhk-dlc-dlk)/4;
for effect of temperature (dhc+dlc-dhk-dlk)/4
and for interaction (dhc-dlc)-(dhk-dlk)?
Thank You in advance for help.
Yours sincerely
Maciej Jończyk
Department of Plant Molecular Ecophysiology
Institute of Plant Experimental Biology
Faculty of Biology, University of Warsaw
02-096 Warszawa, Miecznikowa 1
___________________________________
NOCC, http://nocc.sourceforge.net
--
This email was Anti Virus checked by Astaro Security Gateway. http://www.astaro.com
More information about the Bioconductor
mailing list