[BioC] log Fold Change

[michael watson (IAH-C)]

Assuming your design matrix is c(1,1,-1,-1) then the latter two M values are multiplied by -1:

> mean(c(-1.1907,-1.00507,-0.68244023,-0.815401))
[1] -0.9234028

________________________________________
From: bioconductor-bounces at stat.math.ethz.ch [bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Jay [jboddu at uiuc.edu]
Sent: 16 March 2010 21:29
To: bioconductor at stat.math.ethz.ch
Subject: [BioC] log Fold Change

Hi All:

Even after going through hand full of posts on log fold change, I could not
figure out how exactly it is calculated in LIMMA.

Please see following example.



Arrays 3 and 4 are dye swaps for arrays 1 and 2 (Arrays 1 and 2 are rep 1
and rep 2).



For Gene 1;



The M values are,




1

2

3

4


-1.1907

-1.00507

0.68244023

0.815401



And the A values are,




1

2

3

4


7.693117

6.218356

7.59435237

6.671801



After computing the linear model fit the output is,




logFC

AveExpr

t

P.Value

adj.P.Val

B


-0.9234

7.044406

-8.0598372

0.000229

0.00306

0.868682



I am wondering how the logFC (-0.9234) is gotten.

LIMMA user guide says M values are simply the log2 fold changes between
conditions.



Any help is appreciated.

Thanks

Jay


        [[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor