发信人: leohawk (leohawk), 信区: Biology
标 题: P value !!Re: Also microarray questions
发信站: Unknown Space - 未名空间 (Tue Nov 23 23:23:29 2004) WWW-POST

again, here is some misunderstanding about fold change and P value.

fold change between two chips means nothing, since it could be random or due
to bad probeset or what ever error message.

P value means more, since it is calculated from statistics principle (in most
cases, they, wrongly, assume normal distribution of the gene expression data.
that is why you have to do log transformation, which makes things more like
normal distributed)

a very consistent change, even with small fold change value, may means a lot
to the statistics principle.

So get you mind off the fold change a little bit.

【 在 dentsu (xixi) 的大作中提到: 】
: Thank lots. Actually the data I have in the excel has been processed by
: biostatistician. It was log-transformed before T-test. If I use Bonferroni
: adjustment, only one gene will be shown difference; I will try FDR to see
how
: many genes shown differential expression.My microarray (Affymetrix
drosophila
: chip)was done in university facility. I expected they had pre-processed by
one
: of the two algorithm you mentioned here before gave me the excel
: spreadsheet.Am I right? Another thing I concern is thatI only have 8 genes
: that have both >2 fold change and P<0.05. I was told it is difficult to be
: confirmed if the fold <2. Actaully the top gene (its product is enzyme)I am
: interested has the P value around 10-9, but the fold is 1.26. I tried to
: confrim by measurement of its enzyme activity and had met some problem. I
will
: also try
: ease as you suggested.
: Thank again.
:
:
: 【 在 feir (菲儿) 的大作中提到: 】
: : First of all, if you could find a biostatistician in your department or
your
: : school you can let him/her do the analysis for you. All that you are
asking
: : is standard microarray data analysis.
: :
: : You should use the p-values instead of fold changes in defining
: differentially
: : expressed genes. What puppeteer mentioned (0.05/N) is called Bonferroni
: : adjustment for multiple comparison, which is the most conservative way
: : possible. Other common and less-conservative way is FDR (false-discovery
: : rate) adjustment. Based on my experience, if you have any gene that's
: : significant using Bonferroni you'll have dozens of significant genes using
: : FDR.
: :
: : Another important thing is what algorithm you used to pre-process your
data.
:
: : If you used Affymetrix's MAS 5 you probably want to get rid of all the
genes
: : whose average expression are too low (say 16 or 32) because MAS 5 is not
: great
: : at processing the low-expressed genes so they are likely to be an artifact
: : when they are significant. RMA is a little better on that issue but RMA
: tends
: : to underestimate the fold changes.
: :
: : One more thing you can try is to log-transform your data before you do the
T
: : test comparison, which is almost a standard in microarray analysis. You
may
: : get more signficant genes by doing that because your data would be less
: : skewed.
: :
: : After you get the gene list you can put the probe-set names into EASE
(very
: : simple, just copy-paste), choose the correct chip name in EASE and run it.

: : EASE is a freeware in recgonizing gene themes. Other things you could do
: : include making some pretty clustering pictures, building a prediction
model,
: : matching your list with other lists people have generated before, etc.

--


※ 来源:．Unknown Space - 未名空间 mitbbs.com．[FROM: 68.38.]