The research presented in this PhD thesis employed Next Generation Sequencing (NGS) technologies to investigate phylogenetics, phylogeography, and demography of both extant and extinct deer species. The thesis is divided in two main projects, (i) the use of NGS and ancient DNA to address the aforementioned subjects, and (ii) the development of new molecular methods for reduced representation library sequencing and its applicability to samples stored under sub-optimal conditions. The first section of this thesis is focused on the mitogenomic study of Eurasian deer species during the Late Pleistocene and Holocene using enrichment by capture and shotgun sequencing. In chapter 2, mitogenomes of Megaloceros giganteus, a giant deer species that became extinct around 7,700 calendar years before present in the Urals (Russia), were used to investigate the phylogeography of the species and the effect of the Late Quaternary climatic fluctuations on its extinction. Our results showed the loss of mitochondrial lineages prior to the Last Glacial Maximum (LGM), and demographic estimations identified a decrease in diversity that culminated in the LGM. Stable isotope data was also used to evaluate the dynamics of the species in the Late Quaternary. Overall, both datasets suggested a major climatic influence on the dynamics and extinction of M. giganteus. Chapter 3 presents the first genetic data from two Late Pleistocene Sinomegaceros specimens, an extinct giant deer that lived in east Asia. Here, we generated mitogenome phylogenies that placed Sinomegaceros within the Cervinae family. We also evaluated their evolutionary relationship with other giant deer. The assembled Sinomegaceros mitogenomes were embedded within M. giganteus mitochondrial clade, suggesting mitochondrial introgression from M. giganteus into Sinomegaceros. Chapter 4 consists of two studies analysing pre-LGM mitogenomes of Cervus elaphus and Cervus hanglu as examples of the continuous range fluctuations that occurred in the Late Quaternary within the red deer complex, and their impact on the species’ complex phylogeny and phylogeography. Finally, chapter 5 presents MobiSeq, a novel reduced representation library method to generate genome-wide SNP data. We tested the applicability of MobiSeq in 103 modern and DNA extracts from degraded samples. Our data showed that the method generates a large amount of loci shared among samples with reduced clonality compared to other similar methods that can be used to perform SNP discovery. Our results showed the great potential of MobiSeq and its applicability in a wide range of population genomic studies.