Calculating depth of coverage for overlapping paired-reads

[lre1234]

Hi all,
I have a question on calculating the overall depth of coverage for exome or targeted panel sequencing samples.

As a simple example. If I have paired-end reads that overlap each other. A particular base-pair, would have been sequenced twice, once by the forward read and once by the reverse read. When calculating depth at the base, would you consider this base as having a 1X coverage or a 2X coverage. Essentially treating each individual read of a paired-end as a separate event, or considering them as as combined single event at the base?

Thanks

[SNPsaurus]

I think it depends on the precise definition of depth you are using:
read depth-treat them all independently
fragment depth-count overlaps once
de-duplicated read depth-count overlaps once, and remove PCR duplicates, cluster duplicates, etc.