Are there any research-level applications of proofs by colouring?
This is the kind of proof you use to show that you can't cover a mutilated chessboard with 31 dominoes. Afaik, this technique chiefly finds a market in IMO preparation classes; see Arthur Engel's book or this handout .
Best Answer
I don't think you can formally distinguish "proofs by coloring" from "parity arguments" or more generally arguments that use some mapping into a finite domain, say be reducing mod m. Coloring is just a visually appealing way of looking at such an argument. Needless to say, proofs by parity arguments or other mappings into a finite set are very common; this is a fundamental tool in combinatorics and number theory.