|
发信人: leohawk (leohawk), 信区: Biology
标 题: Re: Also microarray questions
发信站: Unknown Space - 未名空间 (Tue Nov 23 23:30:24 2004) WWW-POST
permutation based test DOES NOT assume normal distribution of data as a lot of
statistics algorithms do.
and it incorporates probability theory into data analysis.
It works by randomly labeling your samples with the categories you do analysis
on, then do a T test (or any test). After doing so many of these things, it
will compared the T test results with the "supposed to be correct" labeling T
test result. and gives you a permutation P value.
by doing so, it actually is the most powerful in eliminating false positives
that may due to "chancy" distribution.
However, there are two main draw backs with this method:
1. it usually requires a lot of results... not a lot people can afford it
2. computationally, it is very expensive, requires a lot CPU time and quite
some memory.
【 在 dentsu (xixi) 的大作中提到: 】
: 【 在 leohawk (leohawk) 的大作中提到: 】
: : 1. what algorithm did you use to calculate the P value? Each method may
: : give a different P value magnitude.
: I use T-test after log-transformation
: :
: : 2. Affymetrix netaffyx has a GO tool, it is kinda useful.
: I will learn more, thanks.
: :
: : 3. It is very hard to publish in good journals with ONLY microarray
: : data now. You need to do more work to confirm your array data. I
: : personally don't belive in PCR confirmation.
: :
: : 4. I would suggest you do a permutation based analysis, one tool you can
: : use is ArrayTools from NCI Biometrix group.
: What is the advantage of this analysis? My problem is that the infection in
: the cell line is very week (actually it is lost three month after use for
: microarray), and the fold change is not very large for those genes that show
: differential expression based on the P value.
:
: Thanks.
: :
: : Good Luck
: :
: : 【 在 dentsu (xixi) 的大作中提到: 】
: : : I had done microarray and gotten the data in excel sheet. My design is
: simple:
: : : comparison of differential gene expression in the infected cell line and
: : : uninfected cell line. I replicate four times for each of them. The
number
: of
: : : genes shown difference is as following:
: : : 1. Based on P value:
: : : 10-9 ~ 10-5: 9 genes
: : : 10-5 ~ 10-4: 32 genes
: : : 10-4 ~ 10-3: 207 genes
: : : 10-2 ~ 0.05: 631 genes
: : : 2. Based on the fold:
: : : >2: 8 genes
: : : 1.5 ~ 2: 40 genes
: : : 1.0 ~ 1.5: 507 genes
: : : My questions are:
: : : 1.Should I use fold or P value to define the differential expression? If
: use P
: : : value, usually what value can be used as threshold? I found someone use
: 10-4.
: : : Is P<0.05 too unstrict for microarray data?
: : : 2.How to classify those differential genes by their function? Should I
use
: : : software to do it or just classify them one by one based on its
: description in
: : : the excel spread sheet? If software, can I input those data directly
from
: the
--
※ 来源:．Unknown Space - 未名空间 mitbbs.com．[FROM: 68.38.]
|
