×The order Squamata is a species-rich group of reptiles including the groups Sauria, Serpentes and Amphisbaenia. Several members of these groups have been extensively used as model organisms for a great variety of studies of different fields including ecology, behaviour and medicine. Such studies require a well-established phylogenetic framework to trace the evolution and diversity of the studied traits in the squamate tree of life. However, the accuracy of phylogenetic inference in Squamata is unbalanced: while the position of the main squamate groups and high level relationships have been successfully and consistently estimated across phylogenetic studies, the phylogenetic relationships of some groups have provided inconsistent results over the years. The accurate estimation of relationships between these groups might have been hindered by methodological artefacts of the phylogenetic inference due to limited data and analytical tools.
For the last couple of decades, and even presently, phylogenetic inference has relied heavily on mitochondrial DNA as the main molecular marker as a result of its advantages, such as the lack of recombination, easy amplification and high evolutionary rate. While its use is based on sound reasons, some drawbacks and limitations have been highlighted, such as the overall small size, and the fact that all genes are linked thus it represents a single locus. To overcome this, the addition of nuclear DNA complements the use of mitochondrial DNA in phylogenetic inference. Within the nuclear DNA, some markers may present so little variation that it is insufficient to recover phylogenetic relationships, yet slow-evolving nuclear genes have been the most widely used markers in phylogenetic studies. On the other hand, it has been shown that the use of highly informative fast evolving nuclear genes, in combination with mitochondrial DNA, can provide resolution at different parts and depths of the evolutionary history of the species.
Phylogenetic analyses of multiple loci have historically been based on the concatenation approach, .i.e. the combination of the sequences into a single alignment that is then analysed to represent the species tree. This method, however, relies on the assumption that the most commonly occurring gene tree is equivalent to the true species tree of any taxa, which may not be true in all cases. Another limitation of the concatenation approach is the elevated percentage of missing data that it sometimes entails, especially in large taxon sets, meaning that the sequence representation is The order Squamata is a species-rich group of reptiles including the groups Sauria, Serpentes and Amphisbaenia. Several members of these groups have been extensively used as model organisms for a great variety of studies of different fields including ecology, behaviour and medicine. Such studies require a well-established phylogenetic framework to trace the evolution and diversity of the studied traits in the squamate tree of life. However, the accuracy of phylogenetic inference in Squamata is unbalanced: while the position of the main squamate groups and high level relationships have been successfully and consistently estimated across phylogenetic studies, the phylogenetic relationships of some groups have provided inconsistent results over the years. The accurate estimation of relationships between these groups might have been hindered by methodological artefacts of the phylogenetic inference due to limited data and analytical tools.
For the last couple of decades, and even presently, phylogenetic inference has relied heavily on mitochondrial DNA as the main molecular marker as a result of its advantages, such as the lack of recombination, easy amplification and high evolutionary rate. While its use is based on sound reasons, some drawbacks and limitations have been highlighted, such as the overall small size, and the fact that all genes are linked thus it represents a single locus. To overcome this, the addition of nuclear DNA complements the use of mitochondrial DNA in phylogenetic inference. Within the nuclear DNA, some markers may present so little variation that it is insufficient to recover phylogenetic relationships, yet slow-evolving nuclear genes have been the most widely used markers in phylogenetic studies. On the other hand, it has been shown that the use of highly informative fast evolving nuclear genes, in combination with mitochondrial DNA, can provide resolution at different parts and depths of the evolutionary history of the species.
Phylogenetic analyses of multiple loci have historically been based on the concatenation approach, .i.e. the combination of the sequences into a single alignment that is then analysed to represent the species tree. This method, however, relies on the assumption that the most commonly occurring gene tree is equivalent to the true species tree of any taxa, which may not be true in all cases. Another limitation of the concatenation approach is the elevated percentage of missing data that it sometimes entails, especially in large taxon sets, meaning that the sequence representation is Miocene and supported a scenario of east-to-west diversification. The little morphological and phylogenetic evidence for the distinctiveness between Rhinechis and Zamenis supports a classification lumping which better reflects their evolutionary history and, based on the priority rule, R. scalaris is moved into the genus Zamenis and designated as Zamenis scalaris comb. nov.
Within the Gallotiinae subfamily, the combined use of fast nuclear markers and the species tree was fruitful for the inference of a robust phylogeny of the genus Psammodromus, distributed in the Iberian Peninsula and North Africa. In this study we analysed for the first time the six Psammodromus species to infer the association between the main cladogenetic events within this genus and the complex biogeographic dynamics across the Strait of Gibraltar. The inferred phylogeographical history suggests that Psammodromus probably originated in Iberia. The African species and the African lineage of P. algirus were a result of two over-sea dispersal events towards Africa that occurred 10 Ma and 1.5 Ma, while continental vicariance events might have shaped the diversification of the species within Iberia and Africa. These results, combined with previous literature, provide compelling evidence that major biotic exchanges occurred across the Strait of Gibraltar well before or long after the land connection during the Messinian Salinity Crisis (5.9 – 5.33 Ma). These findings suggest caution in the application of the relatively short event of the opening of the Strait of Gibraltar at the end of the Messinian Salinity Crisis as a cause for divergence in molecular clock calibrations, which is a common approach in literature.
At the species level, the use of fast evolving nuclear markers proved to be fruitful to understand the geographical pattern of phyletic diversification within the Omanosaura lizards belonging to the Eremiadini tribe, endemic to the Hajar Mountains in the Arabian Peninsula. Multilocus phylogenetic analyses recovered two highly divergent lineages within O. cyanura which are geographically associated to the northernmost and to the south and eastern regions of the Hajar Mountains. These lineages are reciprocally monophyletic at both mitochondrial and nuclear loci suggesting a long history of independent evolution and the need of a comprehensive taxonomic assessment. Omanosaura cyanura represents an additional case of cryptic diversity in the north Hajar Mountains, an area that has been demonstrated to hide high levels of genetic diversity in other reptile groups.
Overall, the work developed in this dissertation has demonstrated that the inclusion of fast evolving nuclear genes and the coalescent species tree approach was relevant for resolving challenging phylogenetic questions on selected squamate groups. This approach allowed the recovery of new clades and provided support for old and recent relationships. These methods have further allowed the comparison between contrasting phylogenetic hypotheses on the Lacertini and Colubrinae, demonstrating that the inference based on the supermatrix approach used by many recent studies may provide high support for incorrect nodes. Future research directions include (i) the application of the phylogenomic approach based on Next Generation Sequence data to further resolve basal polytomy within the Lacertini and Colubrinae radiations; (ii) a taxonomical assessment of Omanosaura; (iii) a re-evaluation of molecular clock calibrations based on the end of the Messinian Salinity Crisis as general cause for divergence, in order to account for biotic exchanges by over-sea dispersal across the Mediterranean well before or long after the land connection during this period, as suggested by emerging literature and the study case on Psammodromus.
