If you know your SQL, you should be able to do queries in BigQuery very quickly. But the native query SQL has some subtle differences to standard SQL (though there is a standard SQL dialect in beta which you can choose to use). So read the documentation so you are confident about the detailed differences.
More importantly, you might have to think differently about how you structure your data. BigQuery likes big denormalized tables: this is almost the opposite of good practice on a relational SQL setup. Performance is usually better with big, flat, denormalized tables than with multiple tables with joins.
Once you have grasped those two ideas, there are few prerequisites. BigQuery is cheap to set up and experiment with and that's the best way to learn.