Summary output plots
Bombcell will output 3 types of plots that can help you get a very quick idea of how well it is performing and if any of the thresholds need to be tweaked.
The plot should look similar to the example one below (but yours will be slightly different!). This plot allows you to quickly see whether the algorithm is classifying noise and non-somatic units correctly.
These plots show the intersections and relationships between multiple sets of data, displaying both the size and composition of these intersections in a compact and easily interpretable format. MOre information on UpSet plots can be found here. Bombcell generates 3: one for units classified as noise, one for units classified as non-somatic and one for units classified as MUA. Looking at them will give you an idea of how well bombcell is performing.
 |
 |
 |
The lines at the bottom indicate how the units are classified based on the metric: red for noise, blue for non-somatic, green for good and orange for MUA.
