Arisen from the Pure Parsimony Haplotyping problem in the bioinformatics, we developed the Minimum Rainbow Subgraph problem (MRS problem): Given a graph $G$, whose edges are coloured with $p$ colours. Find a subgraph $F\\\\subseteq G$ of $G$ of minimum order and with $p$ edges such that each colour occurs exactly once. We proved that this problem is NP-hard, and even APX-hard. Furthermore, we stated upper and lower bounds on the order of such minimum rainbow subgraphs. Several polynomial-time approximation algorithms concerning their approximation ratio and complexity were discussed. Therefore, we used Greedy approaches, or introduced the local colour density $\\\\lcd(T,S)$, giving a ratio on the number of colours and the number of vertices between two subgraphs $S,T\\\\subseteq G$ of $G$. Also, we took a closer look at graphs corresponding to the original haplotyping problem and discussed their special structure.

Arisen from the Pure Parsimony Haplotyping problem in the bioinformatics, we developed the Minimum Rainbow Subgraph problem (MRS problem): Given a graph $G$, whose edges are coloured with $p$ colours. Find a subgraph $F\\\\subseteq G$ of $G$ of minimum order and with $p$ edges such that each colour occurs exactly once. We proved that this problem is NP-hard, and even APX-hard. Furthermore, we stated upper and lower bounds on the order of such minimum rainbow subgraphs. Several polynomial-time approximation algorithms concerning their approximation ratio and complexity were discussed. Therefore, we used Greedy approaches, or introduced the local colour density $\\\\lcd(T,S)$, giving a ratio on the number of colours and the number of vertices between two subgraphs $S,T\\\\subseteq G$ of $G$. Also, we took a closer look at graphs corresponding to the original haplotyping problem and discussed their special structure.:Mathematics and biology - having nothing in common? I. Going for a start 1. Introducing haplotyping 2. Becoming mathematical II. The MRS problem 3. The graph theoretical point of view 3.1. The MRS problem 3.2. The MRS problem on special graph classes 4. Trying to be not that bad 4.1. Greedy approaches 4.2. The local colour density 4.3. MaxNewColour 5. What is real data telling us? And the work goes on and on Bibliography

In this thesis, we deal with rainbow colorings of graphs. We engage not with the rainbow connection number but with counting of rainbow colorings in graphs with k colors. We introduce the rainbow polynomial and prove some results for some special graph classes. Furthermore, we obtain bounds for the rainbow polynomial. In addition, we define some edge colorings related to the rainbow coloring, like the s-rainbow coloring and the 2-rainbow coloring. For this edge colorings, polynomials are defined and we prove some basic properties for this polynomials and present some formulas for the calculation in special graph classes. In addition, we consider in this thesis counting problems related to the rainbow coloring like rainbow pairs and rainbow dependent sets. We introduce polynomials for this counting problems and present some general properties and formulas for special graph classes.

The concept of \emph{proper connection number} of graphs is an extension of proper colouring and is motivated by rainbow connection number of graphs. Let $G$ be an edge-coloured graph. Andrews et al.\cite{Andrews2016} and, independently, Borozan et al.\cite{Borozan2012} introduced the concept of proper connection number as follows: A coloured path $P$ in an edge-coloured graph $G$ is called a \emph{properly coloured path} or more simple \emph{proper path} if two any consecutive edges receive different colours. An edge-coloured graph $G$ is called a \emph{properly connected graph} if every pair of vertices is connected by a proper path. The \emph{proper connection number}, denoted by $pc(G)$, of a connected graph $G$ is the smallest number of colours that are needed in order to make $G$ properly connected. Let $k\geq2$ be an integer. If every two vertices of an edge-coloured graph $G$ are connected by at least $k$ proper paths, then $G$ is said to be a \emph{properly $k$-connected graph}. The \emph{proper $k$-connection number} $pc_k(G)$, introduced by Borozan et al. \cite{Borozan2012}, is the smallest number of colours that are needed in order to make $G$ a properly $k$-connected graph. The aims of this dissertation are to study the proper connection number and the proper 2-connection number of several classes of connected graphs. All the main results are contained in Chapter 4, Chapter 5 and Chapter 6. Since every 2-connected graph has proper connection number at most 3 by Borozan et al. \cite{Borozan2012} and the proper connection number of a connected graph $G$ equals 1 if and only if $G$ is a complete graph by the authors in \cite{Andrews2016, Borozan2012}, our motivation is to characterize 2-connected graphs which have proper connection number 2. First of all, we disprove Conjecture 3 in \cite{Borozan2012} by constructing classes of 2-connected graphs with minimum degree $\delta(G)\geq3$ that have proper connection number 3. Furthermore, we study sufficient conditions in terms of the ratio between the minimum degree and the order of a 2-connected graph $G$ implying that $G$ has proper connection number 2. These results are presented in Chapter 4 of the dissertation. In Chapter 5, we study proper connection number at most 2 of connected graphs in the terms of connectivity and forbidden induced subgraphs $S_{i,j,k}$, where $i,j,k$ are three integers and $0\leq i\leq j\leq k$ (where $S_{i,j,k}$ is the graph consisting of three paths with $i,j$ and $k$ edges having an end-vertex in common). Recently, there are not so many results on the proper $k$-connection number $pc_k(G)$, where $k\geq2$ is an integer. Hence, in Chapter 6, we consider the proper 2-connection number of several classes of connected graphs. We prove a new upper bound for $pc_2(G)$ and determine several classes of connected graphs satisfying $pc_2(G)=2$. Among these are all graphs satisfying the Chv\'tal and Erd\'{o}s condition ($\alpha({G})\leq\kappa(G)$ with two exceptions). We also study the relationship between proper 2-connection number $pc_2(G)$ and proper connection number $pc(G)$ of the Cartesian product of two nontrivial connected graphs. In the last chapter of the dissertation, we propose some open problems of the proper connection number and the proper 2-connection number.