I annotated the vcf file with multiple samples with the help of command line VEP tool of ensembl and the output is:
I just want to know how the same gene location give rise to mutation at two different places in the protein. As we can see that in gene location column we have two same gene positions e.g., first and second row of column "location" we have same position but the mutation in protein amino acid corresponding to that is at different amino acid positions. Also how the amino acid mutation position decreases as the gene location increases along the table.
how the same gene location give rise to mutation at two different places in the protein
What you're calling "same" is just the position on the genome. A single position in the genome can correspond to multiple positions on multiple transcripts - read up on molecular biology concepts such as the central dogma, alternative splicing etc. Note the Feature column - that's the transcript ID. Since these are on different positions in different transcripts, they could also be on different positions in their corresponding protein products. That's why you see 2 different AA changes in the protein change column. If you had protein ID in the annotation, you should see different ENSP IDs as well - they're two different changes in two different protein products, not 2 changes in the same protein product.
how the amino acid mutation position decreases as the gene location increases along the table
"Gene location" (Chromosomal position to be accurate) is assigned to the forward strand of the genome. Transcripts can originate from the forward/positive strand or the reverse/negative strand of the genome. For transcripts originating from the forward strand, AA position will increase with transcript position which will increase with genomic position. For transcripts originating from the reverse strand, AA position will increase with transcript position which will decrease with genomic position. AA position will always change in the same direction as the transcript. Transcript coordinate change along with genomic position change will depend on the strand.
So if I want to find where the mutation in the protein is located I mean if the mutaion is located in some critical motif , which one shouId I use among the same gene positions because the protein residue position is different .
canonical and MANE are not the same transcripts. Canonical = longest. MANE is not necessarily the longest but is the most widely used. See BRCA1 and ESR1 for example.
Canonical might not always be as simple - in my experience it depends a bit on the definition of "canonical" in your group. Ensembl for example state in their canonical genebuild:
We aim to identify the transcript that, on balance, has the highest coverage of conserved exons, highest expression, longest coding sequence and is represented in other key resources, such as NCBI and UniProt.
For human transcripts, the Ensembl Canonical transcript undergoes additional review [...]. Occasionally, this review may result in the designation of a different transcript than the algorithmically selected Ensembl Canonical
In my organization we have yet a different, simpler definition (RefSeq Select if present, longest otherwise), which is closer to yours for example.
So if I want to find where the mutation in the protein is located I mean if the mutation is located in some critical motif , which one should I use among the same gene positions because the protein residue position is different .
So if I want to find where the mutation in the protein is located I mean if the mutaion is located in some critical motif , which one shouId I use among the same gene positions because the protein residue position is different .
There should be domain based annotations (maybe pfam) as well - try and find those.