limma: pvalues for coefficients topTable
1
0
Entering edit mode
Eleni • 0
@605896f8
Last seen 3.2 years ago

Dear members and developers, I have performed a time series experiment with 5 time points on treatment and controls. I used limma linear model for the differential expression analysis. In the design matrix, I included also other parameters (sex, weight and heterophiles/lymphocytes ratio). I find a few significant genes as I see from the top table output (called all coefficients). The code is following:

W2 <- model.matrix(~ Treatment*Time + Sex + dWeight + HL, data=PLSet)
fit <- lmFit(PLSet, W2)
class(fit)
fit <- eBayes(fit)
fit$coefficients[1:5,]
topTable(fit, n=13, coef=2:7)

# corresponding output 
> topTable(fit, n=5, coef=2:7,confint=TRUE)
        TreatmentT         Time        SexM      dWeight            HL TreatmentT.Time  AveExpr
HPS5    4.11164704 -0.067986721 -0.78529467 -0.302848969  0.0043545345     0.076088166 23.71151
PLG.1  -0.46781188  0.002707286 -0.21489308  0.039849887 -0.0011381361     0.048460583 22.32991
AMY2A  -0.66830973 -0.018069170  0.53177497  0.244178510 -0.0061577553     0.106497055 18.74018
MLF1    5.97939600  0.002833836  0.01591788  0.093259276 -0.0151262312    -0.135961018 18.95155
LCP1    0.88863607 -0.025327153 -0.34550418 -0.018652540  0.0040227853     0.003003491 19.74808

               F      P.Value    adj.P.Val
HPS5   18.990758 3.134997e-08 6.395394e-06
PLG.1   9.812850 1.336712e-05 1.363446e-03
AMY2A   8.209525 5.571538e-05 3.788646e-03
MLF1    6.516355 3.040451e-04 1.458770e-02
LCP1    6.365580 3.575417e-04 1.458770e-02

I was wondering: How can I extract the p-value for each coefficient (treatment, time etc. ) for each gene? I know the coefficient column gives an estimate of the importance of the parameter in the result, but is there a way for this estimate to be expressed as a p-value?

Thanks!

TimeSeriesExperiment Regression top limma • 995 views
ADD COMMENT
1
Entering edit mode
@james-w-macdonald-5106
Last seen 1 day ago
United States

You get the data you want by using a single coefficient as an argument to topTable, rather than a range, as you have done. I would strongly suggest you peruse the limma User's Guide, which covers everything in exhaustive detail.

ADD COMMENT

Login before adding your answer.

Traffic: 810 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