<html>
Hi,<br><br>
I tried to start a thread on this in the UCSF microarray list server and
it sort of fell flat, but I like the paper so I'll annoy people here too
:/ <br><br>
Lemon et al. developed an interesting and possibly improved gauge of
confidence called the positive predictive value (PPV) that may be useful
for future scientists looking to test their low level algorithms on known
data sets, but the heart of the paper has to do with their idea on
transforming the intensity values.<br><br>
The authors set out, using a variation on Langmuir's adsorption isotherm
(that is, the classic semi-log sigmoidal dose-response relationship) to
transform the intensity values of individual probes on the array. To me,
this makes more biological sense than some other procedures because it is
based on the ligand-receptor relationship between the probes and the mRNA
species to which they are designed to hybridize.<br><br>
However, when the authors combined their transformed feature level
information into a single measure per probe set, they found that their
procedure (Logit-Exp and Logit-ExpR) performed no better than RMA or
dChip.<br><br>
Interestingly, if they DID NOT collapse their probe level data into a
single probe_set value, and instead tested across all probes (logit-t),
their transformed data did a much better job of winnowing the wheat from
the chaff. They concluded that "...the modeling paradigm may cause
the loss of information from the probe-level data"..<br><br>
This seems critical to me, there is a huge discrepancy between the
significant gene lists generated with different probe level algorithms,
and I don't believe we'll be able to understand why that dichotomy exists
until we look at the underlying probe level information.<br><br>
I have been pleading with our stats department for over a year (I am just
a neuroscientist and I write code like a hippopotamus roller
skates) to employ a 2-way ANOVA on repeated measures at the probe
level to test for significance, and in fact went so far as to put the
notion (with some sample data) into a book chapter I authored earlier
this year (Chapter 6: in A Beginner's Guide to Microarrays).<br><br>
The authors state that "the combination of logit transformation and
probe-level statistical testing provides a means for greatly improved
PPV...". I would agree, but add the caveat that the comparison, at
the probe level, on untransformed values has yet to be done, thus the
probe level idea may be more important than the transformation notion to
improved PPV. Other methods have looked at the probe level information
(e.g., Liu et al. 2002- Affymetrix multiple pairwise comparison- but
their use of the feature level data as biological n may be inapropriate;
and Zhang et al., 2002- but their intention was only for a two chip
comparison).<br><br>
I believe that it is unfortunate that the authors resort to fold change
as a final discriminator after all of that hard work, rather than a
formal statistical test. I still feel that 2-way ANOVA on repeated
measures is the right test for this, but would love to hear from
others.<br><br>
-E<br><br>
P.S. My apologies to Lemon et al if I have misrepresented/ misunderstood
your work. I will gladly retract/ correct this (or any part of it) at
your request.<br><br>
At 12:00 PM 9/25/2003 +0200, you wrote:<br>
<blockquote type=cite class=cite cite>Message: 1<br>
Date: Wed, 24 Sep 2003 19:46:04 +0200<br>
From: "Dario Greco" <greco@biogem.it><br>
Subject: [BioC] ...Logit-t vs RMA...<br>
To: "Bioconductor" <bioconductor@stat.math.ethz.ch><br>
Message-ID: <002601c382c3$bd4a42a0$ce3ca48c@neo><br>
Content-Type:
text/plain;<x-tab> </x-tab>charset="us-ascii"<br><br>
Hi to everybody,<br>
I've just red some days ago the new paper on "logit-t" method
to analyze<br>
affy chips.<br><br>
--------------------------------------------------------------<br>
"A high performance test of differential gene expression for<br>
oligonucleotide arrays"<br><br>
William J Lemon, Sandya Liyanarachchi and Ming You<br><br>
Genome Biology 2003, 4:R67<br>
--------------------------------------------------------------<br><br>
What do you think about this?<br><br>
Regards<br>
Dario<br><br>
--------------------------------------------<br>
Dario Greco<br>
Institute of Genetics and Biophysics <br>
"Adriano Buzzati Traverso" - CNR<br>
111, Via P.Castellino<br>
80131 Naples, Italy<br>
phone +39 081 6132 367<br>
fax +39 081 6132 350 <br>
email: greco@igb.cnr.it; greco@biogem.it</blockquote>
<x-sigsep><p></x-sigsep>
Eric Blalock, PhD<br>
Dept Pharmacology, UKMC<br>
859 323-8033<br><br>
<font size=1><u>STATEMENT OF CONFIDENTIALITY<br><br>
</u></font><font size=1 color="#333333">The contents of this e-mail
message and any attachments are confidential and are intended solely for
addressee. The information may also be legally privileged. This
transmission is sent in trust, for the sole purpose of delivery to the
intended recipient. If you have received this transmission in error, any
use, reproduction or dissemination of this transmission is strictly
prohibited. If you are not the intended recipient, please immediately
notify the sender by reply e-mail or at (859) 323-8033 and delete this
message and its attachments, if any.</font></html>