oalescent-based species delimitation meets deep learning: Insights from a highly fragmented cactus syste

Autor(es)

Perez, Manolo F. ^{[1, 2]} ; Bonatelli, Isabel A. S. ^{[1, 3]} ; Romeiro-Brito, Monique ^[1] ; Franco, Fernando F. ^[1] ; Taylor, Nigel P. ^[4] ; Zappi, Daniela C. ^[5] ; Moraes, Evandro M. ^[1]

Afiliação do(s) autor(es)

^[1] Univ Fed Sao Carlos, Dept Biol, Sorocaba - Brazil

^[2] Univ Fed Sao Carlos, Dept Genet & Evolucao, Sao Carlos - Brazil

^[3] Univ Fed Sao Paulo, Dept Ecol & Biol Evolut, Diadema - Brazil

^[4] Univ Gibraltar, The Alameda - Gibraltar

^[5] Univ Brasilia, Inst Ciencias Biol, Programa Pos Grad Bot, Brasilia, DF - Brazil

• Tipo de documento: Artigo Científico
• Fonte: MOLECULAR ECOLOGY RESOURCES; v. 22, n. 3 OCT 2021.
• Citações Web of Science: 0

Resumo

Delimiting species boundaries is a major goal in evolutionary biology. An increasing volume of literature has focused on the challenges of investigating cryptic diversity within complex evolutionary scenarios of speciation, including gene flow and demographic fluctuations. New methods based on model selection, such as approximate Bayesian computation, approximate likelihoods, and machine learning are promising tools arising in this field. Here, we introduce a framework for species delimitation using the multispecies coalescent model coupled with a deep learning algorithm based on convolutional neural networks (CNNs). We compared this strategy with a similar ABC approach. We applied both methods to test species boundary hypotheses based on current and previous taxonomic delimitations as well as genetic data (sequences from 41 loci) in Pilosocereus aurisetus, a cactus species complex with a sky-island distribution and taxonomic uncertainty. To validate our method, we also applied the same strategy on data from widely accepted species from the genus Drosophila. The results show that our CNN approach has a high capacity to distinguish among the simulated species delimitation scenarios, with higher accuracy than ABC. For the cactus data set, a splitter hypothesis without gene flow showed the highest probability in both CNN and ABC approaches, a result agreeing with previous taxonomic classifications and in line with the sky-island distribution and low dispersal of P. aurisetus. Our results highlight the cryptic diversity within the P. aurisetus complex and show that CNNs are a promising approach for distinguishing complex evolutionary histories, even outperforming the accuracy of other model-based approaches such as ABC. (AU)

• Processo FAPESP: 12/22857-8 - Estrutura genética de espécies de cactaceae do grupo pilosocereus aurisetus utilizando marcadores microssatélites desenvolvidos a partir de sequenciamento de nova geração
• Beneficiário: Isabel Aparecida da Silva Bonatelli
• Modalidade de apoio: Bolsas no Brasil - Doutorado
• Processo FAPESP: 15/06160-5 - Filogenia de Pilosocereus (Cactaceae) e delimitação de espécies no grupo Pilosocereus aurisetus baseado em modelos de coalescência e dados multiloco
• Beneficiário: Evandro Marsola de Moraes
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 12/22943-1 - Filogeografia multilocos do grupo PILOSOCEREUS AURISETUS (Cactaceae)
• Beneficiário: Manolo Fernandez Perez
• Modalidade de apoio: Bolsas no Brasil - Doutorado