Should the sample size be adjusted when performing subsequent analysis using MTAG results? #239
Description
Dear MTAG developer,
Thank you very much for taking the time and effort to answer my questions. When I use MTAG to incorporate six phenotypes, the results output six MTAG phenotypes. For one of them, for example, phenotype A has an original sample size of 10,000 cases + 40,000 controls = 50,000. The mean chi^2 value before MTAG is 1.2, and after MTAG, it is 1.3.
Using the formula to estimate the effective increased sample size due to MTAG:
Effective sample size = 50,000 * [(1 - 1.3) / (1 - 1.2)] = 75,000.
In subsequent analyses (such as TWAS) using the phenotype A-MTAG results, should I use the original sample size of 50,000 or the effective sample size of 75,000?