REQUIRED for all errors and issues:
a) GATK version used: 4.3.0.0
b) Exact command used:  --linked-de-bruijn-graph true --max-reads-per-alignment-start 0

c) Entire program log: run sucess. See detials in the end.

Hi, GATK team! I am analyzing amplicon sequencing data of human mtDNA. A known variant 15607G can not be called in Mutect2 raw vcf result.

I have checked the sam file, the variant indeed exits. Most reads of this variant have flag=0, mapping quality=60, more than 50000x depths, and 100% ALT. I also have variant with higher depths and they can be called correctly. Here is the IGV view of this variant:

As the best practice advised, it's better to set --max-reads-per-alignment-start 0 (to cancel downsampling) for amplicon sequencing data. But I can't call this variant under this setting.

But if I use downsampling and set --downsampling-stride 50, I can call this variant. 

And if I use samtools mpileup instead of Mutect2, I can also call this variant.

Could you please explain why this will happen? 

Besides, some variants seem very strange to me. Two variants have the same PGT and PID. These variants are like this (the GT and AF are marked in red):

How to understand such variant? Should I still filter them based on AF or DP or StrandBias? Or should I filter out them? 

Here are the detail log:

16:00:39.818 INFO  IntelPairHmm - Flush-to-zero (FTZ) is enabled when running PairHMM
16:00:39.820 INFO  IntelPairHmm - Available threads: 8
16:00:39.820 INFO  IntelPairHmm - Requested threads: 4
16:00:39.820 INFO  PairHMM - Using the OpenMP multi-threaded AVX-accelerated native PairHMM implementation
16:00:39.841 INFO  ProgressMeter - Starting traversal
16:00:39.841 INFO  ProgressMeter -        Current Locus  Elapsed Minutes     Regions Processed   Regions/Minute
16:01:00.964 INFO  ProgressMeter -              MT:1490              0.4                    10             28.4
16:01:20.300 INFO  ProgressMeter -              MT:3302              0.7                    20             29.7
16:01:55.966 INFO  ProgressMeter -              MT:5244              1.3                    30             23.6
16:02:15.798 INFO  ProgressMeter -              MT:6992              1.6                    40             25.0
16:02:33.243 INFO  ProgressMeter -              MT:8825              1.9                    50             26.5
16:02:44.049 INFO  ProgressMeter -             MT:10722              2.1                    60             29.0
16:03:06.120 INFO  ProgressMeter -             MT:12548              2.4                    70             28.7
16:03:26.363 INFO  ProgressMeter -             MT:14420              2.8                    80             28.8
16:03:40.293 INFO  ProgressMeter -             MT:16060              3.0                    90             29.9
16:03:40.908 INFO  Mutect2 - 18 read(s) filtered by: MappingQualityReadFilter
0 read(s) filtered by: MappingQualityAvailableReadFilter
0 read(s) filtered by: MappingQualityNotZeroReadFilter
0 read(s) filtered by: MappedReadFilter
0 read(s) filtered by: NotSecondaryAlignmentReadFilter
0 read(s) filtered by: NotDuplicateReadFilter
0 read(s) filtered by: PassesVendorQualityCheckReadFilter
0 read(s) filtered by: NonChimericOriginalAlignmentReadFilter
0 read(s) filtered by: NonZeroReferenceLengthAlignmentReadFilter
0 read(s) filtered by: ReadLengthReadFilter
0 read(s) filtered by: GoodCigarReadFilter
0 read(s) filtered by: WellformedReadFilter
18 total reads filtered
16:03:40.908 INFO  ProgressMeter -             MT:16060              3.0                    92             30.5
16:03:40.908 INFO  ProgressMeter - Traversal complete. Processed 92 total regions in 3.0 minutes.
16:03:40.915 INFO  VectorLoglessPairHMM - Time spent in setup for JNI call : 0.1464474
16:03:40.915 INFO  PairHMM - Total compute time in PairHMM computeLogLikelihoods() : 86.5387548
16:03:40.915 INFO  SmithWatermanAligner - Total compute time in java Smith-Waterman : 11.88 sec
16:03:40.917 INFO  Mutect2 - Shutting down engine
[February 16, 2023 4:03:40 PM CST] org.broadinstitute.hellbender.tools.walkers.mutect.Mutect2 done. Elapsed time: 3.02 minutes.
Runtime.totalMemory()=1944584192
Tool returned:
SUCCESS