I think you should start by the beginning with a general introduction explaining the need to determine your outcome and the way to report specific factors associated with it (OR vs RR...) then the different models using fixed effect vs random effect meta-analysis.

Funnel plot is generally secondary to the main analysis (to look for publication bias).

there are many online ressources but maybe starting with Cochrane website information would be good (https://training.cochrane.org/handbook/current/chapter-10 )

Then many software are available. My preference is R (https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/) but other options are also available.

Hello Anthony, have you been able to get out from it?

Funnel plots are a useful tool for identifying small study effects, such as publication bias, in meta-analyses. They can be created by plotting the effect size of each study included in the meta-analysis against a measure of the study's precision, typically the standard error of the effect size estimate.

To combine data from multiple systematic reviews for creating a funnel plot, you will need to ensure that the studies included in the different reviews are sufficiently similar in terms of the population, intervention, comparator, and outcome measures. Once you have identified the relevant studies, you will need to extract the data on the effect size and the measure of precision for each study.

You will also have to standardize the data so it can be combined, since the studies might have reported their results in different ways. For example, for the effect size, you might have to convert Odds ratios to Risk ratios or Risk differences.

Once you have collected and standardized the data, you can then create a funnel plot by plotting the effect size on the y-axis and the precision (standard error) on the x-axis. You should also include a line that represents the overall effect size estimate from the meta-analysis.

It is also important to note that funnel plots are not conclusive evidence of publication bias, and other methods such as Egger's test or the trim and fill method are recommended to be used as sensitivity analyses.