multtest data preprocessing questions
1
0
Entering edit mode
ulas karaoz ▴ 80
@ulas-karaoz-773
Last seen 9.6 years ago
i am trying to use multest R package(source package) and i am puzzled by the following: 1-What kind of initial processing is done to the data? I thought you get the golub.RData under the /multtest/data directory by running golub.R script that is attached(in the inst/doc directory) but when I run that script on the original data, and do "golub[1:4, 1:4]" i cannot get the same thing as in the documentation, i get: Gene.Accession.Number call call.1 call.2 36 2.000000 2.000000 2.017033 2.000000 37 2.332438 2.064458 2.677607 2.190332 38 2.901458 2.636488 3.168497 2.618048 39 4.162505 2.788875 3.753506 3.685742 2- My other question is, why is the need to center, and to scale the data in the golub.R script, is it ever used? 3-Also in the same script, the class labels are initialized to 1, and 2(i.e. golub.cl<-c(rep(1,27),rep(2,11)) ) although all the statistical tests upstream require 0, or 1 as labels. Am I wrong? Is anyone running into similar problems. thanks.
• 654 views
ADD COMMENT
0
Entering edit mode
ulas karaoz ▴ 80
@ulas-karaoz-773
Last seen 9.6 years ago
i am trying to use multest R package(source package) and i am puzzled by the following: 1-What kind of initial processing is done to the data? I thought you get the golub.RData under the /multtest/data directory by running golub.R script that is attached(in the inst/doc directory) but when I run that script on the original data, and do "golub[1:4, 1:4]" i cannot get the same thing as in the documentation, i get: Gene.Accession.Number call call.1 call.2 36 2.000000 2.000000 2.017033 2.000000 37 2.332438 2.064458 2.677607 2.190332 38 2.901458 2.636488 3.168497 2.618048 39 4.162505 2.788875 3.753506 3.685742 2- My other question is, why is the need to center, and to scale the data in the golub.R script, is it ever used? 3-Also in the same script, the class labels are initialized to 1, and 2(i.e. golub.cl<-c(rep(1,27),rep(2,11)) ) although all the statistical tests upstream require 0, or 1 as labels. Am I wrong? Is anyone running into similar problems. thanks.
ADD COMMENT

Login before adding your answer.

Traffic: 807 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6