number of samples quantified & width violin plots

[Welmoed van Zuiden]

Hello,

In the voila tsv output file (after running heterogen on dozens of human samples), there are columns indicating the number of samples quantified in each condition (condition_num_quantified). I have a few questions related to this:

• Often this number is only a subset of the total number of samples included; is that because these splice events are not detected in all samples, thus only those with detectable PSI levels are included?
• Sometimes no value is specified; does this mean this particular event was detected in all samples included in the heterogen analysis?
• Do these numbers relate to the width of the violin plots that are visualized with voila? Some violin plots are much fuller/wider than others, and I wonder why.

Thank you, best,

Welmoed

[San Jewell]

Hi Welmoed, 

> Often this number is only a subset of the total number of samples included; is that because these splice events are not detected in all samples, thus only those with detectable PSI levels are included

Yes, the condition_num_quantified column specifies the number of samples in that condition which were quantified (and being quantified implies detectable PSI levels)

> Sometimes no value is specified; does this mean this particular event was detected in all samples included in the heterogen analysis?

No, this is unexpected, the code behind this function is simple and should only ever produce an integer zero or greater. If you see nothing at all in this cell, I would be interested in taking a look at your run to see if I can reproduce the result. 

> Do these numbers relate to the width of the violin plots that are visualized with voila? Some violin plots are much fuller/wider than others, and I wonder why.

No, this is also unexpected, in general violin plots maximum width dimension is fixed (though, when the PSI swarm is concentrated, the maximum width part can be quite thin), can you show a screenshot of what is happening?

Thanks!

-San