In this thesis, increased vertebrate genome sampling and recent methodological advancements were combined to address three distinct questions pertaining to vertebrate molecular evolution. Gene duplicability is the tendency to retain multiple gene copies after a duplication event. Various factors correlate with gene duplicability, such as protein function and timing of expression during development. The position of a gene’s encoded product in the protein-protein interaction network recently emerged as an additional factor determining gene duplicability. The first investigation described in this thesis coupled comparative genomics with protein-protein interaction data to assess the dynamic relationship between gene duplicability and network structure in primates. Deciphering the timing of the Ursidae (bear) phylogeny speciation events has proven to be a challenging task. A valuable node to calibrate in such studies is that separating giant panda and polar bear. The exact timing of this important calibration node is currently disputed. The second investigation described in this thesis applied the largest amount of nuclear data currently available in a Bayesian framework to attempt to accurately estimate the timing of divergence between the giant panda and polar bear. It is known that synonymous codon usage is governed by a combination of selective and neutral processes. Currently, it is thought that primarily neutral processes govern synonymous codon usage in vertebrates, possibly due to their lower long-term effective population sizes. The third investigation described in this thesis combined increased genomic sampling and a novel codon usage bias index to conduct the first systematic investigation into the forces that govern synonymous codon usage in vertebrates.