Spanning tree with lower bound on the degrees

TL;DR

This paper provides shorter proofs and polynomial time algorithms for conditions ensuring the existence of spanning trees with prescribed degree lower bounds, with applications to the Weak Nine Dragon Tree Conjecture and planar graph coloring.

Contribution

It introduces new proofs and the first polynomial time algorithm for checking and constructing such spanning trees, improving upon prior theoretical results.

Findings

Provided shorter proofs for existing theorems.

Developed a polynomial time algorithm for the problem.

Applied the results to graph coloring and conjectures.

Abstract

We concentrate on some recent results of Egawa and Ozeki [J. Graph Theory, 2015 and Combinatorica, 2014], and He et al. [J. Graph Theory, 2002]. We give shorter proofs and polynomial time algorithms as well. We present two new proofs for the sufficient condition for having a spanning tree with prescribed lower bounds on the degrees, achieved recently by Egawa and Ozeki. The first one is a natural proof using induction, and the second one is a simple reduction to the theorem of Lov\'asz. Using an algorithm of Frank we show that the condition of the theorem can be checked in time $O(m\sqrt{n})$, and moreover, in the same running time -- if the condition is satisfied -- we can also generate the spanning tree required. This gives the first polynomial time algorithm for this problem. Next we show a nice application of this theorem for the simplest case of the Weak Nine Dragon Tree Conjecture, and for the game coloring number of planar graphs, first discovered by He et al. Finally, we give a shorter proof and a polynomial time algorithm for a good characterization of having a spanning tree with prescribed degree lower bounds, for the special case when $G[S]$ is a cograph, where $S$ is the set of the vertices having degree lower bound prescription at least two. This theorem was proved by Egawa and Ozeki in 2014 while they did not give a polynomial time algorithm.