Question in cancer detection

[cyly]

Hi, I am a beginner of using matlab. I am now doing a project about cancer detection in matlab.
I've got 120 samples which have classified as tumor or normal cells from patients.
For each sample, there is 19200 parameters which are gene information of the cell.
My objective is to find the best combination of parameters to classify the sample whether it is tumor or normal.
Can anyone teach me how to do this? thanks very much

I have already now using t-test in matlab to find some parameters which is useful. however, there are about five hundred and something. Can I further reduce the number of parameters which is useful? By what method?

 

[IRstuff]

Perhaps, you should start with:

http://en.wikipedia.org/wiki/Principal_component

TTFN

FAQ731-376

 

[cyly]

IRstuff, thanks for your reply.
I have a rough idea about PCA and I know that in "Help" in matlab, i can search "PCA". However, it seems there is no result related to my aim. does the "Help" in matlab really help me? or i must write the program by myself to solve the problem?

 

[FeX32]

By what method?
There are many. For example, there are some new techniques using AI predictors such as ANFIS (look it up). However, they need indicies to function, you can't simply shove the data in Matlab.

There is an ANFIF option in the "Fuzzy logic toolbox" that Matlab has. It may save you months of work.
Here is a link (but there are more):

http://www.cs.nthu.edu.tw/~jang/anfisfaq.htm

Fe

 

[FeX32]

Correction in bold:
"There is an ANFIS option in the "Fuzzy logic toolbox" that Matlab has. It may save you months of work. "


Fe