Hi everyone,

how would I build the SDM-tables and input files when calculating a meta-analysis over studies that report two groups (e.g. patients & healthy controls?) I understood from the tutorial that I have to define two columns indicating sample sizes (n1 and n2) and one column  indicating the study name. But then how would the input files with the peak coordinates be named so that the program can associate the peaks with the correct group? Do I need 2 files with peak lists per study?

Best,
Isabel

Dear Isabel, SDM inputs depend on the data you have and the design you want for your meta-analysis:

- If  you have the peaks or maps of the comparisons between groups, then enter each comparison in a row and use "n1" for one group and "n2" for the other group.

- If  you have separate peaks or maps for patients and for controls, then enter each group in a row, use "n", and create a variable to indicate whether the study is in patients or in controls (e.g. the column may be called "patients" and be 1 for patients and 0 for controls). Please note than in this case patients and controls should have been ideally investigated in different studies (although I think it's Ok if there is a reasonable number of patients-controls pairs). Alternatively, you could only include patients and controls studied in the same studies and conduct a paired-measures meta-analysis (that may be difficult), or estimate the patients-controls comparisons from the separate patients and controls maps (this is not straightforward but has been already done).