mrkkk
楼主生物信息狗,做完GWAS,手工计算病例和SNP的OR值,OR值及置信区间的计算公式都不难,但是问题来了:
------------------------------------
allele1 allele2
------------------------------------
case A B
control C D
------------------------------------
OR=AD/BC
95% CI of OR = e^(95% CI of ln(OR))
95% CI of ln(OR) = ln(OR)±1.96(1/A + 1/B + 1/C + 1/D)0.5
如果这四个格子ABCD有一个是0,那如何是好。
有无大神解救下
mrkkk
自己gugu到一点东西,这个Prism adds 是什么意思
When any value equals zero
If any cell has a zero, Prism adds 0.5 to all cells before calculating the relative risk, odds ratio, or P1-P2 (to prevent division by zero).
mrkkk
又找到一个
Alternative empirical continutity corrections (Sweeting et al., 2004) have been introduced to improve calculations where there is a substantial difference between the sizes of the groups being compared. The so called 'reatment arm (sums to 1)' continuity correction performs well with the Mantel-Haenszel pooled estimate. If R is the ratio of treatment to control group sizes then the treatment arm continuity correction for the treatment group is 1/(R+1), and for the control group it is R/(R+1)
libingfei
我想Prism应该是个统计工具名,prism为所有格+0.5。。。。。。.或者使用a/c-b/d代替计算。(防除0)
mrkkk
[未知用户]
这种全加0.5的方法,载case和control的group大小差异很大的情况下,会不会很严重偏离,下面的‘reatment arm (sums to 1)’ continuity correction,你能不能看明白
mrkkk
[未知用户]
貌似第二个方法不是解决这个0的问题的
libingfei
我觉得第二段这个貌似跟你的问题无关。
也可能是我没看懂。
读起来应该是降解决组别间数量差距问题的。
mrkkk
[未知用户]
恩,那是解决数量差异的,那我搞定啦,thanks