How does SNP position numeration work?

Question

[eyb]

Is counting of each SNP position starts from the beginning of each chromosome? Or it is continuous?

For example

In my dataset first snp for chromosome 2 rs2685230 has a position 437664 and last SNP rs10191556 has a position 242521405

But first SNP at chromosome 13 rs11617984 has a position 19622143

Is this a mistake in my data, or it means that 13 chromosome has a first SNP at 19622143? Is it supposed to be shorter than chromosome 2?

Answer 1

[Devon Ryan]

The position is the position within each chromosome. Actually using a continuous numbering system would be a nightmare. Have a look at the VCF spec for further details.

[Farrel]

So on a chromosome where is nucleotide number 0 or number 1. Is it on the very tip of the telomere? Is is the p telomere or the q arm telomere? So does the chromosomal position simply step up by 1 for each nucleotide marching through the centromere to the other arm?

[Devon Ryan]

The first base is at one of the ends (assuming non-circular DNA). Which of the two ends isn't always known. For mouse/human/etc. (i.e., high quality reference genomes), the first base is the tip of the p-arm. Yes, each subsequent base is one position higher.

Answer 2

[Pierre Lindenbaum]

The ID's of the SNP are just a primary key in the dbsnp database (e.g: http://dev.mysql.com/doc/refman/5.0/en/example-auto-increment.html ) . The SNPs occur where the regions have been studied, there is no relationship with the length of the chromosomes. Furthermore, the chr13 is telocentric:

$ curl -s "http://hgdownload.cse.ucsc.edu/goldenPath/hg19/chromosomes/chr2.fa.gz" | gunzip -c | uniq -c | head -n 10
      1 >chr2
    200 NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
      1 CGTATCCcacacaccacacccacacaccacacccacacacacccacaccc
      1 acacccacacacaccacacccacacaccacacccacacccacacaccaca
      1 cccacaccacacccacacaccacacaccacacccacacccacacacacca
      1 cacccacacaccacacccacacacaccctaaccctaacccctaaccccta
      1 accctaaccctacccgaaccctaaccctaaccctaacccctaaccctaac
      1 ccctaaccctaaccctaaccgtaaccctaaccctttaccctaacccgaac
      1 ccctaacccctaacccctaacccttaaccctaacccttaaccctgaccct
      1 gaccctgaccgtgaccctgaccctaacccgaacccgaacccgaaccccga



$ curl -s "http://hgdownload.cse.ucsc.edu/goldenPath/hg19/chromosomes/chr13.fa.gz" | gunzip -c | uniq -c | head -n 10
      1 >chr13
 380400 NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
      1 GAATTCAACTGCCTCCCTGGCCTTTCCCTGCCAATCTATGTGCCCCAGCA
      1 GCCAACTTACATAGCACTGAGTGCAGACTTGTAAATAGACCTTCCAGTTC
      1 TGCTATAATCAATACCTTATTGTCCATAACTCAATTTGGAGAAGGTTTAG
      1 CTGTCTACCAACTCTTGTGGAGAGTTTCTGTGAAGTTTTGTTTTGGGTTG
      1 CAAGAATCTGGAAAACAGATGCAGATGTTTTTGAGGAAGATTTTGAAATT
      1 TCTACTTATAAGGTACCCAAAATGGGATCCAAACTCTTGAATTTGGTTGA
      1 TCTTCTGAAATACATACCTGTGTTTTAAGATTTGCTTGAGCAAACCTTTA
      1 ACCATGGAAATTTTAACCAATGATTTCCAGGTTGAAACAATTCCAGTTTT

so you won't find any SNPs in the 5' region of this chromosome.