bioRxiv Figure 3

[rcannood]

I quite liked Figure 3 from the bioRxiv paper:

I had a few remarks / questions:

• The different plotting styles (pie / circle / square / progressbar) adds no value to the visualization. To me, it seems like the figure would be easier to interpret if the AUROC, AUPRC and Jaccard Index would use the same shape (as was the case in our figures as well).
• I'm missing a summary of the results. According to this Figure, which methods perform well in what areas? This is discussed in the Discussion, but it's not apparent for me in this figure.

[tmmurali]

Hi Robert, Thanks for your interest in the BEELINE manuscript! We have enjoyed reading the Dynverse paper and been inspired by it.

• You are right about the shapes. We are moving to uniform shapes in our figures.

• We are also preparing a summary of the results that highlights the areas in which different methods are strong.

[tmmurali]

Figure 6 in the BEELINE paper in Nature Methods addresses this issue.