Parnassius smintheus SNP and associated weather and landscape variables

Weather is an important short-term, local driver of population size and dispersal, which in turn contribute to patterns of genetic diversity and differentiation within species. Climate change is leading to greater weather variability and more frequent extreme weather events. While the effects of long-term and broad-scale mean climate conditions on genetic variation are well studied, our understanding of the effects of weather variability and extreme conditions on genetic variation is less developed. We assessed the influence of temperature and snow depth on genetic diversity and differentiation of populations of the alpine butterfly, Parnassius smintheus. We examined the relationships between a suite of variables, including those representing extreme conditions, and population-level genetic diversity and differentiation across 1453 single nucleotide polymorphisms, using both linear and gravity models. We additionally examined effects of land cover variables known to influence dispersal and gene flow in this species. We found that extreme low temperature events and the lowest recorded mean snow depth were significant predictors of genetic diversity. Extreme low temperature events, mean snow depth, and land cover resistance were significant predictors of genetic differentiation. These results are congruent with known effects of early winter weather on population size and habitat connectivity on dispersal in P. smintheus. Our results demonstrate the potential for changes in the frequency or magnitude of extreme weather events to alter patterns of genetic diversity and differentiation.

Methods
We sampled 431 Parnassius smintheus whole body vouchers from 21 meadows in Alberta, Canada. DNA was extracted and genotyped using double digest restriction site associated DNA sequencing (ddRADseq). Reads were aligned to the congener Parnassius apollo using Bowtie2. After subsequent filtering (including maximum missing data per SNP of 85%, maximum missing data per individual 50%, minor allele frequency of 0.02), we retained 1453 SNPs for analysis. Expected heterozygosity and chord distance were estimated in R using the package adegenet (Jombart, 2008). 

Temperature variables were obtained from from Natural Resources Canada’s interpolated spatial models (Hutchinson et al., 2009) and estimated over 1960-2014. Snow variables were obtained as 1 km gridded snow depth estimates from the National Operational Hydrologic Remote Sensing Center’s SNODAS model (National Operational Hydrologic Remote Sensing Center, 2004) for 2010-2018.

天气是影响种群规模与扩散的关键短期局地驱动因子，进而作用于物种种内的遗传多样性与遗传分化格局。气候变化正导致天气变异性加剧，极端天气事件发生频次提升。尽管学界已对长期、大尺度的平均气候条件对遗传变异的影响开展了充分研究，但我们对天气变异性与极端条件如何作用于遗传变异的认知仍较为匮乏。本研究以高山蝴蝶红珠绢蝶（Parnassius smintheus）的种群为对象，评估了温度与积雪深度对其遗传多样性及遗传分化的影响。我们采用线性模型与重力模型，基于1453个单核苷酸多态性（single nucleotide polymorphism, SNP）位点，分析了包括极端条件相关变量在内的多类变量与种群水平遗传多样性、遗传分化之间的关联。此外，本研究还探究了已知会影响该物种扩散与基因流的土地覆盖变量的作用。研究结果显示，极端低温事件与实测最低平均积雪深度是遗传多样性的显著预测因子；极端低温事件、平均积雪深度以及土地覆盖阻力则是遗传分化的显著预测因子。上述结果与学界已探明的初冬天气对红珠绢蝶种群规模、生境连通性对扩散的影响相符。本研究结果表明，极端天气事件的频次或强度变化，有可能改变物种种内的遗传多样性与遗传分化格局。

方法
本研究从加拿大阿尔伯塔省的21个草甸中采集了431份红珠绢蝶的全标本作为凭证样本。采用双酶切限制性位点关联DNA测序（double digest restriction site associated DNA sequencing, ddRADseq）技术进行DNA提取与基因分型。使用Bowtie2将测序读段比对至近缘种阿波罗绢蝶（Parnassius apollo）的参考基因组。经后续过滤流程（包括每个单核苷酸多态性位点的最大缺失数据比例为85%、每个个体的最大缺失数据比例为50%、次要等位基因频率阈值为0.02）后，最终保留1453个SNP位点用于后续分析。使用R语言的adegenet包（Jombart, 2008）估算期望杂合度与弦距离。

温度变量源自加拿大自然资源部的插值空间模型（Hutchinson等, 2009），其时间范围为1960-2014年。积雪变量则取自美国国家业务水文遥感中心的SNODAS模型（National Operational Hydrologic Remote Sensing Center, 2004）的1 km分辨率格网积雪深度估算数据，时间范围为2010-2018年。