Variants in pseudogene or repeat region

Hi all,

We used PCR to enrich the target amplicon, and get sequenced on HiSeq X.
In our targets, several are located in pseudogenes or repeat region.

We used bwa mem to align the reads to genome with default settings.
BQSR is neglected due to time consuming.

UG is applied to call given variants according to discussions in forum.
However, reads belong to pseudogenes or repeat region can be mapped to other regions, with XA tag in bam file.
This leads to low or even 0 mapping quality for target amplicons, and therefore, the variants are not called.

I searched the forum, and found that changing the Kmer size might fix this, but increase the false positives.

My question is,

1. Shall I use HC instead?
2. What can I do to get variants in these regions called, as well as those 'normal' ones.

Thanks