Question

Defining the design matrix

0

Entering edit mode

humberto_munoz • 0

@humberto_munoz-10903

Last seen 8.4 years ago

I am defining my design matrix with the following results below.

> RG<-calcNormFactors(RG, method=c("TMM"),
+ refColumn=1, logratioTrim=.3, sumTrim=0.05, doWeighting=TRUE,
+ Acutoff=-1e10, p=0.75)
> RG$samples
files group lib.size norm.factors
ARef1 ARef1.csv 1 2911652 1.1218729
ARef24 ARef24.csv 1 2911652 1.1218729
CO21 CO21.csv 2 11198918 1.0289838
CO224 CO224.csv 2 11294616 0.7995743
Light1 Light1.csv 3 12454627 0.7104398
Light24 Light24.csv 3 8668044 0.7049407
NaCl1 NaCl1.csv 4 6550232 0.7554182
NaCl24 NaCl24.csv 4 11475570 0.8774054
NaNO31 NaNO31.csv 5 10521142 1.3035997
NaNO324 NaNO324.csv 5 9045259 1.4339691
pH1 pH1.csv 6 11850666 1.0481779
pH24 pH24.csv 6 9275757 0.7866712
Temp1 Temp1.csv 7 11726477 1.2511421
Temp24 Temp24.csv 7 8120977 1.5085468
> # plotMDS(RG)
> # nrow(RG)
> Condition<- factor(c(ARef,ARef,CO2,CO2,Light,Light,NaCl,NaCl,NaNO3,NaNO3,pH,pH,Temp,Temp))
Error in factor(c(ARef, ARef, CO2, CO2, Light, Light, NaCl, NaCl, NaNO3, :
object 'ARef' not found
> Time<- factor(c("1","24", "1", "24", "1", "24", "1", "24", "1", "24","1", "24", "1", "24"))
> data.frame(Sample=colnames(RG),Condition,Time)
Error in data.frame(Sample = colnames(RG), Condition, Time) :
object 'Condition' not found I do not understand why the object ARef is not found when currently is appearing in RG. I will appreciate any solution to this issue.

Thanks

design matrix • 1.0k views

ADD COMMENT • link 8.4 years ago humberto_munoz • 0

0

Entering edit mode

The following lines are executed very well.

Condition<-factor(c("ARef","ARef","CO2","CO2","Light","Light","NaCl","NaCl","NaNO3","NaNO3","pH","pH","Temp","Temp"))

Time<- factor(c("1", "24","1", "24", "1", "24", "1", "24", "1", "24","1", "24", "1", "24"))
data.frame(Sample=colnames(RG),Condition,Time)
design <-model.matrix(~~Condition+Time)
rownames(design)<-colnames(RG)
design

----------------------------------------

Now, I would like to create the following plots:

1) The # of false discoveries vs the # of genes selected for the Fisher test (total reads) and the Fisher test (TMM)

2) The # of false discoveries vs the # of genes selected for the Poisson-Exact and Poisson-LR-TMM normalized with my datasets.

Any helpful to perform these plots will be appreciated.

ADD REPLY • link updated 8.3 years ago by Gordon Smyth 52k • written 8.4 years ago by humberto_munoz • 0

score 1 · Answer 1 · 2016-07-16

1

Entering edit mode

Aaron Lun ★ 28k

@alun

Last seen 2 hours ago

The city by the bay

You need to read up on how R works. Your Condition assignment is looking for a variable named ARef, which doesn't exist in the current session. What you really want is the string "ARef", so you should be doing:

Condition <- factor(c("ARef","ARef","CO2","CO2","Light","Light",
    "NaCl","NaCl","NaNO3","NaNO3","pH","pH","Temp","Temp"))

... or even easier:

Condition <- factor(sub("(1|24)$", "", rownames(RGsamples)))

Obviously, once Condition is defined, then the second error will also be resolved.