| Summary: | 2 Specific goal 1. Study the performance of the NSGA-II optimising four criteria: minimum evolution, least squares, maximum parsimony and likelihood. TOTAL We modified our previous published MO-MA inference algorithm that optimises parsimony and likelihood to include the minimum evolution and least-squares as criteria. We used classical data sets from the related literature to evaluate the algorithm and performed applied case studies using our data related to yeast strains. We formalised the results in the related publications (10.1109/CIBCB48159.2020.9277700; 10.1016/j.biosystems.2022.104606). Additionally, we adapted this algorithm to successfully deal with classification and feature selection problems testing new operators in the field of healthcare management (10.1109/SCCC51225.2020.9281282; 10.1109/SCCC54552.2021.9650434). 3 Specific goal 2. Design of new strategies to infer phylogenetic trees based on the current modifications proposed by the literature for the NSGA-II to treat many-objective optimisation problems. TOTAL To improve MO-MA's performance treating large data sets, we replaced its topological operations based on graphs with matrix operations. It allows us the inclusion of additional inference criteria without increasing the consumption time. Thanks to that, we designed advanced algorithms, including multi-modal and NSGA-III operations, that were presented in the published manuscripts. 4 Specific goal 3. Characterise different crossover and mutation strategies applied to the many-objective phylogenetic inference problem, and its effect on the search process. TOTAL We built and tested different operators to design our multi- and many-optimisations inference algorithms: nearest neighbour interchange, subtree prune and regraft, and tree bisection and reconnection moves. Also, we tested different topological metrics to study the decision space. Additionally, we evaluated the performance of the prune-deleted and other crossover operators. We included the results in our mentioned publications. 5 ...
|
|---|