paired-end adapted mean coverage is only one-third of mean coverageData

[吳德倫（Soappp）]

Hello! I used `qualimap bam qc -ip -gff <bed>` to measure my bam and found that paired-end adapted mean coverage is only one-third of mean coverageData. In my understanding, even if every reads pair are overlapping, the non-overlapping coverage also won't be lower than half of mean coverage. May you let me know how paired-end adapted mean coverage is calculated? I would like to check if is there something I misunderstanding.

this is my result:

>>>>>>> Reference

     number of bases = 3,137,454,505 bp
     number of contigs = 86


>>>>>>> Globals

     number of windows = 485

     number of reads = 5,675,603
     number of mapped reads = 5,673,555 (99.96%)
     number of secondary alignments = 0

     number of mapped paired reads (first in pair) = 2,837,289
     number of mapped paired reads (second in pair) = 2,836,266
     number of mapped paired reads (both in pair) = 5,671,507
     number of mapped paired reads (singletons) = 2,048
     number of overlapping read pairs = 342,399

     number of mapped bases = 64,336,925 bp
     number of sequenced bases = 64,326,452 bp
     number of aligned bases = 0 bp
     number of duplicated reads (flagged) = 487,523


>>>>>>> Globals inside

     regions size = 7,371 (0%)
     number of mapped reads = 755,298 (13.31%)
     number of mapped paired reads (first in pair) = 378,308
     number of mapped paired reads (second in pair) = 376,990
     number of mapped paired reads (both in pair) = 754,355
     number of mapped paired reads (singletons) = 943


>>>>>>> Insert size

     mean insert size = 19,551.7206
     std insert size = 1,232,016.0351
     median insert size = 161


>>>>>>> Mapping quality

     mean mapping quality = 59.0415


>>>>>>> ACTG content

     number of A's = 17,491,278 bp (27.19%)
     number of C's = 14,500,340 bp (22.54%)
     number of T's = 17,956,543 bp (27.91%)
     number of G's = 14,351,451 bp (22.31%)
     number of N's = 26,830 bp (0.04%)

     GC percentage = 44.85%


>>>>>>> Mismatches and indels

    general error rate = 0.0028
    number of mismatches = 322,792
    number of insertions = 2,511
    mapped reads with insertion percentage = 0.04%
    number of deletions = 6,983
    mapped reads with deletion percentage = 0.12%
    homopolymer indels = 57.48%


>>>>>>> Coverage

     mean coverageData = 8,728.3849X
     std coverageData = 2,187.9531X
     paired-end adapted mean coverage = 2,993.1X