I have 10 GEO datasets each for diabetes and cancer and want to find common genes between them. Essentially, I'm trying to replicate the protocol from this study: https://www.nature.com/articles/s41598-023-49715-9 . I'm a beginner in bioinformatics, and I'm confused about how they normalised multiple datasets using GEO2r.

I also heard that I need to Combine Datasets and Perform Batch Effect Correction? is this necessary? does this mean I need to individually analyse the datasets with GEO2r and then download the results, followed by combining?

I will be using performing additional filtering from the results based on p value, fold change etc. followed by GSEA just like the study, and hence need the resulting datasets to be comparable.

the R code used by the study can be found here: https://github.com/RajveerSingh27R/Cross-Phenotype-Analysis/blob/main/DEG_Hub_Gene_R_Code.R