Finding the tree of life is one of the major challenges that scientists are attempting to solve. It is widely believed that the evolution of species can (mostly) be depicted in a tree graph, the phylogenetic tree. However, the true phylogenetic species tree is often unknown. One approach is to computationally infer phylogenetic trees from phylogenetic information encoded in genomic data. With the advancement of sequencing techniques, we have a rapidly growing availability of phylogenetic data, which enable the construction of large-scale phylogenetic trees. This thesis addresses algorithmic issues for the construction of large-scale phylogenetic species trees, the supertrees, and the exploration and analysis of large-scale phylogenetic trees. We present (i) new algorithms for local search methods for supertree construction that reduce the time complexity by an order of magnitude and a parallelization for these methods, (ii) new methods for constructing better supertrees from estimated trees and inferring small, exact phylogenetic trees, and (iii) a novel, interactive visual method for the large-scale tree exploration and the concurrent analysis of multiple gene trees and one species tree.