Landscape complexity influences patterns of animal dispersal, which in turn may affect both gene flow and the spread of pathogens. White-nose syndrome (WNS) is an introduced fungal disease that has spread rapidly throughout eastern North America, causing massive mortality in bat populations. We tested for a relationship between the population genetic structure of the most common host, the little brown myotis (Myotis lucifugus), and the geographic spread of WNS to date by evaluating logistic regression models of WNS risk among hibernating colonies in eastern North America. We hypothesized that risk of WNS to susceptible host colonies should increase with both geographic proximity and genetic similarity, reflecting historical connectivity, to infected colonies. Consistent with this hypothesis, inclusion of genetic distance between infected and susceptible colonies significantly improved models of disease spread, capturing heterogeneity in the spatial expansion of WNS despite low levels of genetic differentiation among eastern populations. Expanding our genetic analysis to the continental range of little brown myotis reveals strongly contrasting patterns of population structure between eastern and western North America. Genetic structure increases markedly moving westward into the northern Great Plains, beyond the current distribution of WNS. In western North America, genetic differentiation of geographically proximate populations often exceeds levels observed across the entire eastern region, suggesting infrequent and/or locally restricted dispersal, and thus relatively limited opportunities for pathogen introduction in western North America. Taken together, our analyses suggest a possibly slower future rate of spread of the WNS pathogen, at least as mediated by little brown myotis.
Myotis cytb parsimony tree
Parsimony tree used to create Myotis lucifugus haplotype network, including outgroups and other Myotis lucifugus sequences downloaded from GenBank. See Key to cytb haplotypes for identification of individual sequences belonging to each haplotype.
cytb.parsimonyGB.tree
Key to cytb haplotypes
Sample ID and/or GenBank Accession number of sequences used to generate the cytb parsimony tree. The corresponding haplotype ID used in the tree file, and number of identical sequences belonging to the haplotype are provided.
CytbTreeKey.txt
Myotis cytb haplotype sequences
Cytb haplotype sequences (fasta format) used to create haplotype network. See Key to cytb haplotypes for samples and GenBank accession numbers belonging to each haplotype.
myluPars.fasta
SNPs from range-wide Myotis lucifugus ddRAD-seq
SNPs (STRUCTURE format) from ddRAD-seq data used in principal component analysis of range-wide samples. The first two columns give the sample name (with allele number) and the sampling locality. Headers for the remaining columns indicate the RAD locus and SNP number. -9 indicates missing data; 0 and 1 indicate indels; 2,3,4,5 indicate A,C,G,T, respectively
RADalln5010_84507.stru
ddRAD-seq from hibernating Myotis lucifugus colonies
ddRAD-seq data (nexus format) used to estimate PhiST among hibernating colonies of Myotis lucifugus for risk models of disease spread. Sequences represent unphased, concatenated RAD sequences. Headers indicate the sample name and locality. Only loci with genotype calls for all samples were included.
EriskCmplt871RVAPPAL.nex
ddRAD-seq from range-wide Myotis lucifugus
ddRAD-seq data (fasta format) from Myotis lucifugus sampled across their range. Unphased sequences were concatenated for each individual and used to estimate PhiST across sampling localities. Headers indicate sample ID (with allele number) and sampling locality. Only loci with genotype calls for all samples were included.
Perfect_Clusters.fasta
Cytb sequences from hibernating Myotis lucifugus colonies
Cytb sequences (fasta format) from hibernating colonies of Myotis lucifugus used to estimate PhiST for risk models of disease spread. Header indicates sample ID and sampling locality.
Cytb_Erisk.fasta
Genetic and geographic distance matrix from range-wide ddRAD-seq
Pairwise PhiST of ddRAD-seq (above diagonal) and geographic distance (km; below diagonal) between localities of Myotis lucifugus sampled across their range. The first four columns indicate Region (E=eastern, W=western), Site-type (M=maternity colony, H=hibernaculum, F=foraging site), sample size (N) and sampling locality (Site).
RAD_Rangewide.txt
Genetic and geographic distance matrix from range-wide cytb
Pairwise PhiST of cytb (above diagonal) and geographic distance (km; below diagonal) between localities of Myotis lucifugus sampled across their range. The first four columns indicate Region (E=eastern, W=western), Site-type (M=maternity colony, H=hibernaculum, F=foraging site), sample size (N) and sampling locality (Site).
Cytb_Rangewide.txt
Genetic and geographic distance matrix for ddRAD-seq among hibernacula
Pairwise PhiST of ddRAD-seq (above diagonal) and geographic distance (km; below diagonal) between hibernating colonies of Myotis lucifugus used in risk models of disease spread. Column 1 gives sample size (N) and column 2 gives sampling locality (Site).
RAD_Erisk.txt
Genetic and geographic distance matrix for cytb among hibernacula
Pairwise PhiST of cytb (above diagonal) and geographic distance (km; below diagonal) between hibernating colonies of Myotis lucifugus used in risk models of disease spread. Column 1 gives sample size (N) and column 2 gives sampling locality (Site).
Cytb_Erisk.txt