[BioC] Volcano plots

[James MacDonald]

Volcano plots are useful for visualizing differentially expressed genes
that have already been detected using a statistical test. They are a
plot of the log fold change on the x-axis and some sort of positive
quantity on the y-axis. Often the y-axis is -log10(p-value) where the
p-value comes from an F-test. In the limma vignette the y-axis is the
log odds.

So the volcano plot is not used to detect differentially expressed
genes, but to make a pretty picture showing the differentially expressed
genes. They have also been used to show that large fold change does not
necessarily equal statistical significance.

Jim



James W. MacDonald
UMCCC Microarray Core Facility
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623

>>> mathob at gimr.garvan.unsw.edu.au 06/24/03 07:23PM >>>
I too have been wondering about the usefulness of the volcano plot and
so was 
interested in this thread.  But (excuse my slowness) what's the answer
to the 
original  question about whether they can be used to help detect 
differentially expressed genes?

Matthew


> recently I was asked about volcano plots. I have the idea that they
> are used for detection of differentially expressed genes, but I am
not
>
> quite certain.

-- 
----------------------------------------------------------------------
Matthew Hobbs

Garvan Institute of Medical Research
384 Victoria St            Ph   : (02) 9295 8327
Darlinghurst                
http://www.garvan.org.au        email:  m.hobbs at garvan.org.au 

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch 
https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor