g**********y 发帖数: 423 | 1 我得到这个数据
CellTypeI CellTypeII CellTypeIII
Hypermethylated 5400 1600 500
Hypomethylated 450 1500 150
用Fisher exact test or chisq test只能说明 methyaltion和CellType是相关的。
但是不能表明CellType中的哪种enriched with hyper or hypo-methylation。
显然,CellTypeI中Hypermethylated的数目比较多,要用什么test来说明
CellTypeI is significantly enriched with Hypermethylation。
另外 Is CellTypeII enriched with Hypomethylation?
Thanks!!! |
P****D 发帖数: 11146 | 2 Don't quite understand your question. Perhaps you want to calculate Odds
Ratio? |
s*****9 发帖数: 108 | 3 顶楼上,lz想比较不同的cell type之间的差异吧,用odds-ratio呗。 |
g**********y 发帖数: 423 | |
g**********y 发帖数: 423 | 5 个人感觉odds-ratio不行,因为odds-ratio要求binary value (p1+p2=1)
实际上这个表还有一行
CellTypeI CellTypeII CellTypeIII
Hypermethylated 5400 1600 500
Hypomethylated 450 1500 150
non-diff to ref 14150 16900 19350
就是说column sum == 20000 |
g**********y 发帖数: 423 | |
g**********y 发帖数: 423 | 7 或者permutation test or hypergeometric test on each row? |
j******4 发帖数: 6090 | 8 整理成这样再做exact test?
CellTypeI other type |
y****2 发帖数: 65 | 9 armitage trend test?
http://www.statgen.org/main/images/www_statgen_org/downloads/Da
score test (logistic regression, odd ratio)可以用么? |