de novo vs somatic caller Varscan

Question

[ccscience76]

Hello

Im trying to figure out to use different tools to study de novo variants from whole exome sequence.

If I merge the parental bam files(marking duplicates later and so on) and I use somatic variant caller from Varscan against the child, the results shouldn't be a bit similar if I use directly de novo(trio) caller tool from Var Scan?

Maybe this sounds a bit weird...but I would appreciate if someone can explain to me.

Thanks