This thesis presents a novel approach to reconstructing ancestral genomes of a number of descendant species related by a phylogeny. Traditional methods face challenges due to cycles of whole genome doubling followed by fractionation in plant lineages. In response, the thesis proposes a new approach that first accumulates a large number of candidate gene adjacencies specific to each ancestor in a phylogeny. A subset of these which to produces long ancestral contigs are chosen through maximum weight matching. The strategy results in more complete reconstructions than existing methods, and a number of quality measures are deployed to assess the results. The thesis also presents a new computational technique for estimating the ancestral monoploid number of chromosomes, involving a "g-mer" analysis to resolve a bias due to long contigs and gap statistics to estimate the number. The method is applied to a set of phylogenetically related descendant species, and the monoploid number is found to be 9 for all rosid and asterid orders. Additionally, the thesis demonstrates that this result is not an artifact of the method, by deriving a monoploid number of approximately 20 for the metazoan ancestor. The reconstructed ancestral genomes are functionally annotated and visualized through painting ancestral projections on descendant genomes and highlighting syntenic ancestor-descendant relationships. The proposed method is applied to genomes drawn from a broad range of plant orders. The Raccroche pipeline reconstructs ancestral gene orders and chromosomal contents of the ancestral genomes at all internal vertices of a phylogenetic tree, and constructs chromosomes by counting the frequencies of ancestral contig co-occurrence on the extant genomes, clustering these for each ancestor, and ordering them. Overall, this thesis presents a significant contribution to the field of ancestral genome reconstruction, offering a new approach that produces more complete reconstructions and provides valuable insights into the evolutionary process giving rise to the gene content and order of extant genomes.

The matching of a tournament could be a tedious task, especially if there are many teams takingpart. To combat this difficult task this thesis evaluates if, by creating a method, there is a wayto automate this matching. The method proposed is based on only two parameters, shortestdistance and whether the teams have faced each other already. By being based on shortestdistance the method could not only shorten the planning time but also shorten the distance thatthe participants have to travel.To test the proposed method and validate whether it would give a sufficiently good matching itis applied to an already played Bridge tournament. The tournament was played in 2019 andincluded 365 teams and 7 rounds. This tournament was matched manually based partly ondistance and whether the teams have faced each other. Teams were split into clusters andrandomly matched each other within those. Thus, the teams did not often play against the closestteam but never a team very far away. Results are given in the form of maps for each round anda table which compares the distances for both the methods.The method was applied successfully, and each round was, as excepted, a shorter distance thanthe original matching. Thus, proving that an automatic tournament matching can be done. Thereare some improvement and alteration that can be done in the future to find a matching that morelifelike.