Landscape genomics of a widely distributed snake (Dolichophis caspius, Gmelin, 1789) across Eastern Europe and Western Asia

Across the distribution of the Caspian whipsnake (Dolichophis caspius), populations have become increasingly disconnected due to habitat alterations. To understand population dynamics and the adaptive potential of this widespread but locally endangered snake, we investigated population structure, admixture and effective migration patterns. We took a landscape-genomic approach to identify selected genotypes associated with environmental variables relevant to D. caspius. With double-digest restriction-site associated DNA (ddRAD) sequencing of 53 samples resulting in 17,518 single nucleotide polymorphisms (SNPs), we identified eight clusters within D. caspius reflecting complex evolutionary patterns of the species. Estimated Effective Migration Surfaces (EEMS) revealed higher-than-average gene flow in most of the Balkan Peninsula and lower-than-average gene flow along the middle section of the Danube River. Landscape genomic analysis identified 751 selected genotypes correlated with seven climatic variables. Isothermality correlated with the highest number of selected genotypes (478) located in 41 genes, followed by annual range (127) and annual mean temperature (87). We conclude that environmental variables, especially the day-to-night temperature oscillation in comparison to the summer-to-winter oscillation, may have an important role in the distribution and adaptation of D. caspius.Funding provided by: Nemzeti Kutatási Fejlesztési és Innovációs HivatalCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100011019Award Number: 2017-2.2.4-TÉT-AT-2017-00002Funding provided by: Austrian Science FundCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100002428Award Number: P29623-B25Funding provided by: Agentúra na Podporu Výskumu a VývojaCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100005357Award Number: APVV-15-0147Funding provided by: Nemzeti Kutatási Fejlesztési és Innovációs HivatalCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100011019Award Number: K-124375Funding provided by: YounInvestigators Programme (FiKut) of the Hungarian Academy of Sciences (MTA)*Crossref Funder Registry ID: Award Number: Funding provided by: Agentúra na Podporu Výskumu a VývojaCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100005357Award Number: APVV-19-0076Funding provided by: Austrian Agency for International Cooperation in Education and Research*Crossref Funder Registry ID: Award Number: HU 02/2018Funding provided by: YounInvestigators Programme (FiKut) of the Hungarian Academy of Sciences (MTA)Crossref Funder Registry ID: Funding provided by: Austrian Agency for International Cooperation in Education and ResearchCrossref Funder Registry ID: Award Number: HU 02/2018One hundred twenty-four DNA samples extracted from carcass tissue, shed skin and blood and buccal swabs originating from eight European and Asian countries were collected for this study and deposited at the Collection of Genetic Resources of the Hungarian Natural History Museum in Budapest. Spatial data (longitude and latitude in decimal degrees) were recorded using a GPS device for each sampled individual (Figure 1, Table 1). We extracted DNA from buccal swabs using the blackPREP Swab DNA Kit according to kit protocols (Analytik Jena AG, Jena, Germany). For shed skin and carcass (liver and muscle) samples, the DNEasy Blood and Tissue Kit (Qiagen, Hilden, Germany) was applied following manufacturer's instructions but adding 30 μl of Proteinase K (20 mg/μl) to increase lysis efficacy. Prior to the extraction, skin samples were soaked in water for 24 hours at room temperature and manually sliced into fragments of ~5mm to facilitate enzymatic lysis. Samples' DNA quality (measured by absorbance) was determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, USA). The 96 best extractions were selected based on a concentration threshold of 10 ng/μL and 260/280 ratio values > 1.85 and used for ddRAD sequencing