Plink Meta-Analysis Data Without Standard Error Column (Se)
2
2
Entering edit mode
8.9 years ago
lsvijfhuizen ▴ 90

Dear all,

I would like to run the basic command for the meta-analysis as follow:

plink --meta-analysis study1.assoc study2.assoc


• SNP SNP idenitifier
• OR Odds ratio (or BETA, etc)
• SE Standard error of OR (or user-defined weight field)
• P (Optional) p-value from test

My problem is that file 2 (study2.assoc) does not contain SE (standard error) numbers. File study2.assoc contains three columns named SNP, OR and P. Is there a way to calculate the SE based on these three columns? The three columns is the only information that I have.

I hope that someone can help me with this problem!

Groetjes Lisanne

analysis plink error • 7.6k views
1
Entering edit mode
8.9 years ago
zx8754 10k

How to calculate Standard Error of Odds Ratios? http://stats.stackexchange.com/questions/10375/how-to-calculate-standard-error-of-odds-ratios

0
Entering edit mode
7.4 years ago

Specify the --ci 0.95 command to your --linear or --assoc or --logistic tests, then you'll get SE column along with 95% confidence intervals.