Specific case that ABRA2 is not realigning correctly

[parsaeskandar]

Hi, I am using ABRA2 on a bam file, and it is working well in most cases, but there is position chr19:20618690-20618698, which realigning is not working correctly. I am looking for an explanation for why this case is not realigning as it should.

Here is the IGV snapshot from this region,

Here is the command I am using with the ABRA2:
java -jar $ABRA2 --targets region.bed --in HG001.chr19.left_shifted.indel_realigned.bam --out HG001.chr19.region_realigned.bam --ref GRCh38_GIABv3_no_alt_analysis_set_maskedGRC_decoys_MAP2K3_KMT2C_KCNJ18.fasta --index --threads 16

The .bed file I am using contains chr19 20618650 20618790 and I am using ABRA2 v2.23.

Let me know what other information I can provide! Thank you!