Excessive Amount of Bisulfite Ambiguous

[pablovargasmejia]

I'm testing different bisulfite mappers for my data. BSBolt completely outperforms Bismark (Mappability 94% vs. 35%). However, 70M of 71M of my reads are "Bisulfite Ambiguous". Are the default mapping parameters too lax? Could this amount of ambiguous reads be because my isolate is somewhat different from the reference genome?

bwa mem -Y -A 1 -B 4 -D 0.5 -E 1,1 -L 30,30 -T 10 -U 17 -W 0 -c 500 -d 100 -k 19 -m 50 -r 1.5 -t 20 -w 100 -y 20 -O 6,6 -h 100,200 -e 0.1 -l 0.5 -n 5 -Z 0.95 genome/LT15-INDEX/BSB_ref.fa C1_r1.fq.gz C1_r2.fq.gz

Alignment Complete: Time 4:43:38

Total Reads: 71455594
Mappability: 94.301 %

Reads Mapped to Watson_C2T: 33763063
Reads Mapped to Crick_C2T: 33619994
Reads Mapped to Watson_G2A: 33769331
Reads Mapped to Crick_G2A: 33613726

Unmapped Reads (Single / Paired Ends): 8145074
Bisulfite Ambiguous: 70779193

I would appreciate any insights or suggestions on how to address this issue.