skip to main content


Title: Improving the taxonomy of fossil pollen using convolutional neural networks and superresolution microscopy

Taxonomic resolution is a major challenge in palynology, largely limiting the ecological and evolutionary interpretations possible with deep-time fossil pollen data. We present an approach for fossil pollen analysis that uses optical superresolution microscopy and machine learning to create a quantitative and higher throughput workflow for producing palynological identifications and hypotheses of biological affinity. We developed three convolutional neural network (CNN) classification models: maximum projection (MPM), multislice (MSM), and fused (FM). We trained the models on the pollen of 16 genera of the legume tribe Amherstieae, and then used these models to constrain the biological classifications of 48 fossilStriatopollisspecimens from the Paleocene, Eocene, and Miocene of western Africa and northern South America. All models achieved average accuracies of 83 to 90% in the classification of the extant genera, and the majority of fossil identifications (86%) showed consensus among at least two of the three models. Our fossil identifications support the paleobiogeographic hypothesis that Amherstieae originated in Paleocene Africa and dispersed to South America during the Paleocene-Eocene Thermal Maximum (56 Ma). They also raise the possibility that at least three Amherstieae genera (Crudia,Berlinia, andAnthonotha) may have diverged earlier in the Cenozoic than predicted by molecular phylogenies.

 
more » « less
NSF-PAR ID:
10199101
Author(s) / Creator(s):
; ; ; ; ; ; ;
Publisher / Repository:
Proceedings of the National Academy of Sciences
Date Published:
Journal Name:
Proceedings of the National Academy of Sciences
Volume:
117
Issue:
45
ISSN:
0027-8424
Page Range / eLocation ID:
p. 28496-28505
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Hyaenodonta is a diverse, extinct group of carnivorous mammals that included weasel- to rhinoceros-sized species. The oldest-known hyaenodont fossils are from the middle Paleocene of North Africa and the antiquity of the group in Afro-Arabia led to the hypothesis that it originated there and dispersed to Asia, Europe, and North America. Here we describe two new hyaenodont species based on the oldest hyaenodont cranial specimens known from Afro-Arabia. The material was collected from the latest Eocene Locality 41 (L-41, ∼34 Ma) in the Fayum Depression, Egypt.Akhnatenavus nefertiticyonsp. nov. has specialized, hypercarnivorous molars and an elongate cranial vault. InA. nefertiticyonthe tallest, piercing cusp on M1–M2is the paracone.Brychotherium ephalmosgen. et sp. nov. has more generalized molars that retain the metacone and complex talonids. InB. ephalmosthe tallest, piercing cusp on M1–M2is the metacone. We incorporate this new material into a series of phylogenetic analyses using a character-taxon matrix that includes novel dental, cranial, and postcranial characters, and samples extensively from the global record of the group. The phylogenetic analysis includes the first application of Bayesian methods to hyaenodont relationships.B. ephalmosis consistently placed within Teratodontinae, an Afro-Arabian clade with several generalist and hypercarnivorous forms, andAkhnatenavusis consistently recovered in Hyainailourinae as part of an Afro-Arabian radiation. The phylogenetic results suggest that hypercarnivory evolved independently three times within Hyaenodonta: in Teratodontinae, in Hyainailourinae, and in Hyaenodontinae. Teratodontines are consistently placed in a close relationship with Hyainailouridae (Hyainailourinae + Apterodontinae) to the exclusion of “proviverrines,” hyaenodontines, and several North American clades, and we propose that the superfamily Hyainailouroidea be used to describe this relationship. Using the topologies recovered from each phylogenetic method, we reconstructed the biogeographic history of Hyaenodonta using parsimony optimization (PO), likelihood optimization (LO), and Bayesian Binary Markov chain Monte Carlo (MCMC) to examine support for the Afro-Arabian origin of Hyaenodonta. Across all analyses, we found that Hyaenodonta most likely originated in Europe, rather than Afro-Arabia. The clade is estimated by tip-dating analysis to have undergone a rapid radiation in the Late Cretaceous and Paleocene; a radiation currently not documented by fossil evidence. During the Paleocene, lineages are reconstructed as dispersing to Asia, Afro-Arabia, and North America. The place of origin of Hyainailouroidea is likely Afro-Arabia according to the Bayesian topologies but it is ambiguous using parsimony. All topologies support the constituent clades–Hyainailourinae, Apterodontinae, and Teratodontinae–as Afro-Arabian and tip-dating estimates that each clade is established in Afro-Arabia by the middle Eocene.

     
    more » « less
  2. Premise

    Solanaceae is a scientifically and economically important angiosperm family with a minimal fossil record and an intriguing early evolutionary history. Here, we report a newly discovered fossil lantern fruit with a suite of features characteristic of Physalideae within Solanaceae. The fossil comes from the early Eocene Laguna del Hunco site (ca. 52 Ma) in Chubut, Argentina, which previously yielded the only other physaloid fruit fossil,Physalis infinemundi.

    Methods

    The fruit morphology and calyx venation pattern of the new fossil were compared withP. infinemundiand extant species of Solanaceae.

    Results

    Physalis hunickeniisp. nov. is clearly distinct fromP. infinemundiin its fruiting calyx with wider primary veins, longer and thinner lobes, and especially in its venation pattern with high density, transverse tertiary veins; these features support its placement in a new species. In comparison with extant physaloid genera, the calyx venation pattern and other diagnostic traits reinforce placement of the new fossil, likeP. infinemundi, within the tribe Physalideae of Solanaceae.

    Conclusions

    Both species of fossil nightshades from Laguna del Hunco represent crown‐group Solanaceae but are older than all prior age estimates of the family. Although at least 20 transoceanic dispersals have been proposed as the driver of range expansion of Solanaceae, the Patagonian fossils push back the diversification of the family to Gondwanan times. Thus, overland dispersal across Gondwana is now a likely scenario for at least some biogeographic patterns, in light of the ancient trans‐Antarctic land connections between South America and Australia.

     
    more » « less
  3. Abstract

    Nyssa(Nyssaceae, Cornales) represents a classical example of the well‐known eastern Asian–eastern North American floristic disjunction. The genus consists of three species in eastern Asia, four species in eastern North America, and one species in Central America. Species of the genus are ecologically important trees in eastern North American and eastern Asian forests. The distribution of living species and a rich fossil record of the genus make it an excellent model for understanding the origin and evolution of the eastern Asian–eastern North American floristic disjunction. However, despite the small number of species, relationships within the genus have remained unclear and have not been elucidated using a molecular approach. Here, we integrate data from 48 nuclear genes, fossils, morphology, and ecological niche to resolve species relationships, elucidate its biogeographical history, and investigate the evolution of morphology and ecological niches, aiming at a better understanding of the well‐known EA–ENA floristic disjunction. Results showed that the Central American (CAM)Nyssa talamancanawas sister to the remaining species, which were divided among three, rapidly diversified subclades. Estimated divergence times and biogeographical history suggested thatNyssahad an ancestral range in Eurasia and western North America in the late Paleocene. The rapid diversification occurred in the early Eocene, followed by multiple dispersals between and within the Erasian and North American continents. The genus experienced two major episodes of extinction in the early Oligocene and end of Neogene, respectively. The Central AmericanN. talamancanarepresents a relic lineage of the boreotropical flora in the Paleocene/Eocene boundary that once diversified in western North America. The results supported the importance of both the North Atlantic land bridge and the Bering land bridge (BLB) for the Paleogene dispersals ofNyssaand the Neogene dispersals, respectively, as well as the role of Central America as refugia of the Paleogene flora. The total‐evidence‐based dated phylogeny suggested that the pattern of macroevolution ofNyssacoincided with paleoclimatic changes. We found a number of evolutionary changes in morphology (including wood anatomy and leaf traits) and ecological niches (precipitation and temperature) between the EA–ENA disjunct, supporting the ecological selection driving trait evolutions after geographic isolation. We also demonstrated challenges in phylogenomic studies of lineages with rapid diversification histories. The concatenation of gene data can lead to inference of strongly supported relationships incongruent with the species tree. However, conflicts in gene genealogies did not seem to impose a strong effect on divergence time dating in our case. Furthermore, we demonstrated that rapid diversification events may not be recovered in the divergence time dating analysis using BEAST if critical fossil constraints of the relevant nodes are not available. Our study provides an example of complex bidirectional exchanges of plants between Eurasia and North America in the Paleogene, but “out of Asia” migrations in the Neogene, to explain the present disjunct distribution ofNyssain EA and ENA.

     
    more » « less
  4. Premise

    Eocene floras of Patagonia document biotic response to the final separation of Gondwana. The conifer genusAraucaria, distributed worldwide during the Mesozoic, has a disjunct extant distribution between South America and Australasia. Fossils assigned to AustralasianAraucariaSect.Eutactausually are represented by isolated organs, making diagnosis difficult.Araucaria pichileufensisE.W. Berry, from the middle Eocene Río Pichileufú (RP) site in Argentine Patagonia, was originally placed in Sect.Eutactaand later reported from the early Eocene Laguna del Hunco (LH) locality. However, the relationship ofA. pichileufensisto Sect.Eutactaand the conspecificity of theAraucariamaterial among these Patagonian floras have not been tested using modern methods.

    Methods

    We review the type material ofA. pichileufensisalongside large (n= 192) new fossil collections ofAraucariafromLHandRP, including multi‐organ preservation of leafy branches, ovuliferous complexes, and pollen cones. We use a total evidence phylogenetic analysis to analyze relationships of the fossils to Sect.Eutacta.

    Results

    We describeAraucaria huncoensissp. nov. fromLHand improve the whole‐plant concept forAraucaria pichileufensisfromRP. The two species respectively resolve in the crown and stem of Sect.Eutacta.

    Conclusions

    Our results confirm the presence and indicate the survival of Sect.Eutactain South America during early Antarctic separation. The exceptionally complete fossils significantly predate several molecular age estimates for crownEutacta. The differentiation of twoAraucariaspecies demonstrates conifer turnover during climate change and initial South American isolation from the early to middle Eocene.

     
    more » « less
  5. Abstract

    Skippers are a species rich and widespread group of butterflies with evolutionary patterns and processes largely unstudied despite some recent efforts. Among Hesperiidae, the subfamily Heteropterinae is a moderately diverse clade comprising ca. 200 species distributed from North to South America and from Africa to the Palearctic region. While some regions are species rich, others are far less diverse. Using anchored phylogenomics, we infer a robust timetree and estimate ancestral ranges to understand the biogeographic history of these skippers. Inferences based on up to 383 exons recover a robust backbone for the subfamily along with the monophyly of all genera. Bayesian divergence time estimates suggest an origin of Heteropterinae in the late Eocene, ca. 40 million years ago. Maximum likelihood ancestral range estimates indicate an origin of the group in the New World. The eastern Palearctic was likely colonized via a Beringian route and a reverse colonization event resulted in two independent and extant American clades. We estimate a vicariant event between Central and South America that significantly predates estimates of the proto‐Caribbean seaway closure, indicating active overwater dispersal in the Oligocene. The colonization of Africa from the east Palearctic is synchronous with the closure of the Tethys Ocean, while the colonization of Madagascar appears to be comparatively recent. Our results shed light on the systematics and biogeography of Heteropterinae skippers and unveil the evolutionary history of a new leaf in the skipper tree‐of‐life.

     
    more » « less