Understanding coverage in VCF

[elfuser]

Hello and sorry if this question has been asked before or it is super obvious.
I'm having a bit of a hard time finding out what the numbers mean in the vcf pileup.
I created an alignment BAM file of my illumina paired-end reads to a bacterial genome using the following:

bwa aln ref.fa r1.fq > r1.sai
bwa aln ref.fa r2.fq > r2.sai
bwa sampe ref.fa r1.sai r2.sai r1.fq r2.fq > aln.sam
samtools view -bS aln.sam > aln.bam

Then I create the pileup with

samtools mpileup -uf aln.bam > pile.bcf
bcftools view pile.bcf > pile.vcf

I also create a depth file with

samtools depth aln.bam > depth.txt

Let's take position 214 in the reference genome
In the vcf

gi|218888746|ref|NC_011770.1|	214	.	G	C	202	.	DP=45;AF1=1;AC1=2;DP4=0,0,25,13;MQ=55;FQ=-141	GT:PL:GQ	1/1:235,114,0:99

In the depth

gi|218888746|ref|NC_011770.1|	214	53

I understand that the DP(45) value in the vcf file refers to the raw coverage of the given position with high quality(q13?). An that the coverage number in the depth file (53) refers to the number of bases called at that given position regardless of quality and/or conflict. I also understand that the DP4 values (0,0,25,13) reflect if the reads that cover that position match the reference or an alternative variant. In this case 38 (25+13) high quality reads match the alternative variant and 0 the reference. If this is right what happened to the other 7 (45-38) high quality reads that cover that position.

Another doubt I have is when I find multiple genotypes at a position. For example:
In the VCF

gi|218888746|ref|NC_011770.1|	3081	.	C	T,G	213	.	DP=70;AF1=1;AC1=2;DP4=0,0,31,32;MQ=50;FQ=-214	GT:PL:GQ	1/1:246,187,0,247,170,244:99

My understanding is that in this case that 63 reads support either a T or a G at this position. So what I am trying to understand is how I can tell the support for each genotype. I tried looking at the GT:PL:GQ, but I dont quite understand it.
Thanks!