The ability to assign biological samples to source populations based on genetic variation with high accuracy and precision is important for numerous applications from ecological studies through wildlife conservation to epidemiology. However, population assignment when genetic differentiation is low is challenging, and methods to address this problem are lacking. The application of artificial neural networks to population assignment using genomic data is highly promising. Here we present popfinder: a new, easy to use Python-based artificial neural network pipeline for genetic population assignment. We tested popfinder both with simulated genetic data from populations connected by varying levels of gene flow, and with reduced-representation sequence data for three species of seabirds with weak to no population genetic structure. Popfinder was able to assign individuals to their source populations with high accuracy, precision and recall in most cases, including both simulated and empirical datasets, except in the weakest empirical population structure dataset, where the comparator programs also performed poorly. Compared to other available software, popfinder was slower on the simulated data sets due to hyperparameter tuning and the fact that it does not reduce the dimensionality of the data set; however, all programs ran in seconds on empirical data sets. Additionally, popfinder provides a perturbation ranking method to help develop optimized SNP panels for genetic population assignment, and is designed to be user-friendly. Finally, we caution users of all assignment programs to watch both for leakage of data during model training, and for unequal detection probabilities.

The relative contributions of past and present evolutionary processes in shaping population genetic differentiation can be difficult to ascertain, especially in highly mobile animals. The Northern Gannet (Morus bassanus; hereafter, gannet) is a migratory colonial seabird that is widely distributed across the North Atlantic Ocean. Despite strong dispersal abilities, decades of banding and tracking studies indicate that the North Atlantic is a barrier for seasonal migration of gannets: gannets breeding in North America winter along the southeastern coast of USA and Mexico, while European breeders winter along the western European and African coasts. However, telemetry recently revealed that some gannets migrate across the ocean, suggesting that trans-Atlantic gene flow is possible. We investigated patterns of genomic variation among gannets from 12 colonies across the species’ range using double digest restriction-site associated sequencing (ddRADseq). Indices of genetic differentiation, principal component analysis, a Bayesian clustering method, and discriminant analysis of principal components all indicated that gannets breeding in North America versus Europe differ genetically, in accordance with segregation at both breeding and non-breeding areas. However, Bayesian assignment methods indicated that low, unidirectional introgression occurs from Europe into North America, suggesting that the North Atlantic is a semi-permeable barrier to gene flow in gannets. Evolutionary modeling suggested that the two genetic populations originated in separate refugia during the Pleistocene and underwent secondary contact during the Holocene. These results are consistent with results of previous studies and provide direct evidence that seasonal migratory behaviour can influence population genetic structure in a highly mobile organism.

The mechanisms that restrict gene flow between populations and facilitate population differentiation and speciation vary across the tree of life. In systems where physical barriers to gene flow are dynamic over time and space, such as many marine species, introgression may be a major factor in the speciation process. In sympatric species of seabirds with no land barriers between them, hybridization has been frequently observed but few studies have investigated patterns of introgression. We used whole-genome sequence data to test for interspecific introgression between five pairs of tropical and subtropical seabirds and to test for gene flow within species across major land masses and ocean basins. We found evidence for introgression between blue-footed (Sula nebouxii) and Peruvian boobies (S. variegata); masked (S. dactylatra) and Nazca boobies (S. granti); and blue-footed and Nazca boobies. We found no evidence of introgression between blue-footed and brown boobies (S. leucogaster) and masked and brown boobies despite observed hybridization between these species. We also found evidence for gene flow across several major land masses in three pantropical species: red-footed (S. sula), brown, and masked boobies. Finally, we report evidence for ancient introgression between brown boobies and the ancestor of blue-footed, Peruvian, masked, and Nazca boobies. Our work indicates (1) that interspecific introgression has shaped contemporary booby diversity in the eastern Pacific, and (2) that contemporary physical barriers to gene flow between booby colonies are not impenetrable. Our findings contribute novel insights to the growing body of evidence that suggests introgression is a widespread evolutionary process.