Reduced coverage in vcf

[bariceta]

I am analyzing quite short amplicons (<1kb) with ONT and even in my bam file (aligned with epi2me/wf-alignment) I can see a coverage around 1000000X, but after variant calling using NanoCaller last version available the depth of a known SNP is around 10000X (100 times lower than expected).

Should I change any parameter to obtain the real depth? I change the maxcov parameter to 1000000 to try to increase the value, but I always obtain the same depth.

Thanks