I am learning microarray meta-analysis and heterogenity is a really essential test after we carry out meta-analysis. However, there are some points that I still do not understand:
1. There are 3 ways to test heterogenity: Q-value, I-squared and tau-squared. Do we only check the result of only 1 of these 3 values in 1 heterogenity test or do we check all of them?
2. What is the popular code for R that people now use for heterogenity test? I have checked that we have metaQC package but now, it has been removed. So, are there any function or package that we can use? And the statistical method in metaQC are quite different to the 3 -way heterogenity method that I listed before( Qvalue, I-squared and tau-squared). I remember that they use IQC, EQC (internal/external quality control) value. I just wonder if those value is another index to check heterogenity of meta-analysis (which is used for the same purpose as q-value, i-squared) or it is used for different purpose?
3. There are cases that heterogenity can be inside the dataset (between the probes) and the case that heterogenity can be among datasets. Are the method to test heterogenity between these 2 cases difference or the same? If I just work for microarray meta-analysis, do I need to test the heterogenity inside dataset?
Thank you for your attention and please help me to understand those.
Yours sincerely.
Suggested reference
Integrated analysis of the heterogeneous microarray data.
Article Integrated analysis of the heterogeneous microarray data
Dear Aijit Kumar Roy,
Thank you so much for your suggestion but the heterogenity that I meant in my question is about the differences between dataset (which dataset is outlier?) and inside dataset (between samples, which samples is outlier?). So can you suggest me any method that I should do?
- Badges
- Science topic