Computational Drug Discovery and Design

3.5 Subgroup
Analysis (SeeNote 8)

MiRNAs are differentially expressed among tissue types and spe-

cies, with corresponding overall effects and heterogeneities. Sub-

group analyses divide the extracted data of miRNA expression

profiling according to tissue types (blood, muscle, pancreas, liver,

etc.) and species (i.e., tissue specificity and species specificity).

Technical details and full examples could be found in [7].

3.6 Sensitivity
Analysis

Sensitivity analysis is performed on using stricter inclusion criteria

such as sample sizes to test the robustness of findings. Sample size is

a dominant factor affecting precision in determining the overall

effects. Technical details and full examples could be found in [7].

4 Notes

1. RStudio is easier to use than ordinary R environment. RStudio
   will work only after proper installation of R.


2. When you have a question of interest suitable for conducting a
   meta-analysis, you should check whether there are existing
   meta-analyses, whether the existing meta-analyses can
   fully answer your question, and how your meta-analysis
   would make a difference in providing stronger evidence.


3. Detailed search strategy and search dates for databases should
   be documented for reproducibility and updating. You may
   follow the method to update the results.


4. Figure3 shows the example results from a PubMed search, you
   should not only update it but also add the search results from
   other databases. The selection criteria are often based on the

Fig. 8R codes used for meta-analysis

MicroRNA Meta-analysis 481