Low novel Ti/Tv ratio in expanded exome data

Hey!
I'm using VSQR to filter variants from expanded exome data including exome (RocheNimblegen v3, 64Mb), HLA region (Roche Nimblegen
HLA, 5 Mb) and selection of regulatory regions (64 Mb) captured using a custom designed SeqCap kit (Roche NimbleGen). I have combined data from 90 samples and TS-filter level for both SNPs and indels has been 99.0. However, when analysing filtered data with VariantEval tool, the novel Ti/Tv ratio is 0.77 (known dbsnp_138.b37 2.15 and total 1.88). With this kind of data should I use some extra arguments when running VSQR?
Best regards,
Kati Hyvärinen
FRCBS, Finland