setting up the correlation matrix with two channel, dye swapped data . How to deal with Bio and Tech replicates?
0
0
Entering edit mode
@richard-green-4701
Last seen 10.2 years ago
Howdy folks, I could use some help. I have 32 two arrays that I need to run differential expression on. My target file (with notes) is attached. The comparisons are : infected_time_2h mock_2h infected_time_6h mock_6h infected_time_24h mock_24h mock_2h infected_time_2h mock_6h infected_time_6h mock_24h infected_time_24h The arrays are dye swapped. Basically I would like to compare my infected time point with my mock time point. But here is the catch, I have pairs of biological replicates and then there were sets of technical replicates done off of those,so my question is how to address the biololgical and technical replicates within the design and how to address viewing these differences within a contrast matrix . Would you set it up this way and average the dye swapped with those that are not dye swapped? Is this still taking into account the technical replicates and calculating differential expression? It appears to be. design <- cbind(INF2hvsMOCK2h = c(1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), INF2hvsMOCK2hneg = c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,-1,-1,-1,-1,-1,0,0,0,0,0,0,0,0,0, 0), INF6hvsMOCK = c(0,0,0,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), INF6hvsMOCKneg = c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,-1,-1,-1,0,0,0,0,0,0) , INF24hvsMOCK = c(0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), INF24hvsMOCKneg = c(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,-1,-1,-1,-1,- 1)) cont.matrix <- makeContrasts(INF2hvsMOCK2h_avg = (INF2hvsMOCK2h + INF2hvsMOCK2hneg)/2, INF6hvsMOCK6h_avg = (INF6hvsMOCK + INF6hvsMOCKneg)/2, INF24hvsMOCK24h_avg = (INF24hvsMOCK + INF24hvsMOCKneg)/2, levels = design) fit2 <- contrasts.fit(fit, cont.matrix) Or this way? biolrep <- c(1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,-1,-1,-2,-2,-3,-3,-4,-4,-5,-5,-6,-6, -7,-7,-8,-8) library("statmod") corfit <- duplicateCorrelation(E2.avg, ndups = 1, block = biolrep) fit <- lmFit(E2.avg, block = biolrep, cor = corfit$consensus) This does seem to calculate DE but takes into account the replicates This way does not account for the contrast and looking at differences Any advice folks have is much appreciated. Thanks -Rich -------------- next part -------------- Filename Cy3 Cy5 biological replicate technical replicate US23502338_251485025519_S01_GE2_107_Sep09_1_3.txt time_2h mock_2h 1 1 US23502338_251485025519_S01_GE2_107_Sep09_1_4.txt time_2h mock_2h 2 1 US23502338_251485025520_S01_GE2_107_Sep09_1_3.txt time_2h mock_2h 1 2 US23502338_251485025520_S01_GE2_107_Sep09_1_4.txt time_2h mock_2h 2 2 US23502338_251485025521_S01_GE2_107_Sep09_1_3.txt time_2h mock_2h 1 3 US23502338_251485025521_S01_GE2_107_Sep09_1_4.txt time_2h mock_2h 2 3 US23502338_251485029145_S01_GE2_107_Sep09_1_3.txt time_6h mock_6h 1 1 US23502338_251485029145_S01_GE2_107_Sep09_1_4.txt time_6h mock_6h 2 1 US23502338_251485029151_S01_GE2_107_Sep09_1_3.txt time_6h mock_6h 1 2 US23502338_251485029151_S01_GE2_107_Sep09_1_4.txt time_6h mock_6h 2 2 US23502338_251485029633_S01_GE2_107_Sep09_1_3.txt time_24h mock_24h 1 1 US23502338_251485029633_S01_GE2_107_Sep09_1_4.txt time_24h mock_24h 2 1 US23502338_251485029602_S01_GE2_107_Sep09_1_3.txt time_24h mock_24h 1 2 US23502338_251485029602_S01_GE2_107_Sep09_1_4.txt time_24h mock_24h 2 2 US23502338_251485029141_S01_GE2_107_Sep09_1_3.txt time_24h mock_24h 1 3 US23502338_251485029141_S01_GE2_107_Sep09_1_4.txt time_24h mock_24h 2 3 US23502338_251485025519_S01_GE2_107_Sep09_1_1.txt mock_2h time_2h -1 -1 US23502338_251485025519_S01_GE2_107_Sep09_1_2.txt mock_2h time_2h -2 -1 US23502338_251485025520_S01_GE2_107_Sep09_1_1.txt mock_2h time_2h -1 -2 US23502338_251485025520_S01_GE2_107_Sep09_1_2.txt mock_2h time_2h -2 -2 US23502338_251485025521_S01_GE2_107_Sep09_1_1.txt mock_2h time_2h -1 -3 US23502338_251485025521_S01_GE2_107_Sep09_1_2.txt mock_2h time_2h -2 -3 US23502338_251485029145_S01_GE2_107_Sep09_1_1.txt mock_6h time_6h -1 -1 US23502338_251485029145_S01_GE2_107_Sep09_1_2.txt mock_6h time_6h -2 -1 US23502338_251485029151_S01_GE2_107_Sep09_1_1.txt mock_6h time_6h -1 -2 US23502338_251485029151_S01_GE2_107_Sep09_1_2.txt mock_6h time_6h -2 -2 US23502338_251485029633_S01_GE2_107_Sep09_1_1.txt mock_24h time_24h -1 -1 US23502338_251485029633_S01_GE2_107_Sep09_1_2.txt mock_24h time_24h -2 -1 US23502338_251485029602_S01_GE2_107_Sep09_1_1.txt mock_24h time_24h -1 -2 US23502338_251485029602_S01_GE2_107_Sep09_1_2.txt mock_24h time_24h -2 -2 US23502338_251485029141_S01_GE2_107_Sep09_1_1.txt mock_24h time_24h -1 -3 US23502338_251485029141_S01_GE2_107_Sep09_1_2.txt mock_24h time_24h -2 -3
• 884 views
ADD COMMENT

Login before adding your answer.

Traffic: 633 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6