Joining Clinvar with variant data using sql

Question

[iCode]

I want to join my variant data (from 23andMe) with clinvar and I see people Join them differently. Specifically, these are the join on clauses

            sample.chromosome = clinvar.chromosome
    AND sample.startposition = clinvar.startposition - 1
    AND sample.endposition = clinvar.endposition
    AND sample.referenceallele = clinvar.referenceallele

that was used in this blog: https://aws.amazon.com/blogs/big-data/interactive-analysis-of-genomic-datasets-using-amazon-athena/

As you see, clinvar.startposition has a minus 1 (clinvar.startposition - 1) but I am seeing others are not doing so and just say

  ...
  AND sample.startposition = clinvar.startposition
  ...

What is the proper way to join my variants with clinvar. should I have that minus 1 on the startposition or not? and why?

[Pierre Lindenbaum]

it depends how data have been inserted. Cheat Sheet For One-Based Vs Zero-Based Coordinate Systems

[iCode]

Thank you. Would you have happened to know what Coordinate Systems 23andMe api data and Clinvar data are in?