Bryophytes; convolutional neural networks; scanning electron microscopy; species identification; spores; sporoderm; ultrastructure
Abstract :
[en] [en] BACKGROUND AND AIMS: Automatized species identification tools have massively facilitated plant identification. In mosses, spore ultrastructure appears to be a promising taxonomic character, but has been largely under-exploited. Here, we test artificial intelligence-based approaches to identify species from their spore morphology. In particular, we determine whether the number of spores, their polarity, and variation among populations and capsules affect model accuracy.
METHODS: Scanning electron microscopy spore images were generated for five capsules of five populations in ten species. Convolutional neural networks with a highly modularized architecture (ResNeXt) were trained to identify the species, population and capsule of origin of a spore. The training set was progressively sub-sampled to test the impact of sample size on model accuracy. To assess whether variation in spore morphology among populations affected model accuracy, one population was successively removed to test a model trained on the four remaining populations.
KEY RESULTS: Species were correctly identified at average rates of 92 %, regardless of polarity. Model accuracy decreased progressively with decreasing sample size, dropping to about 80 % with 15 % of the initial dataset. The population and capsule of origin of a spore was retrieved at rates >75 %, indicating the presence of diagnostic population and capsule markers on the sporoderm. Strong population structure in some species caused a substantial drop of model accuracy when model training and testing was performed on different populations.
CONCLUSIONS: Spore morphology appears to be an extremely promising tool for moss species identification and may usefully complement the suite of morphological characters used so far in moss taxonomy. The presence of spore diagnostic features at the population and capsule level raises substantial questions on the origin of this structure, which are discussed. Substantial infraspecific variation makes it necessary, however, to train an automatized identification tool from a range of populations and capsules.
Milis, Alix ✱; Université de Liège - ULiège > Integrative Biological Sciences (InBioS) ; Research and Collection Departments, Botanic Garden Meise, Nieuwelaan 38, Meise B-1860, Belgium
Hofmann, Martin ✱; Fakultät für Informatik und Automatisierung, Data-Intensive Systems and Visualization, Technische Universität Ilmenau, Ehrenbergstraße 29, Ilmenau 98693, Germany
Mäder, Patrick ; Fakultät für Informatik und Automatisierung, Data-Intensive Systems and Visualization, Technische Universität Ilmenau, Ehrenbergstraße 29, Ilmenau 98693, Germany ; German Centre for Integrative Biodiversity Research, iDiv (Halle-Jena-Leipzig), Puschstraße 4, Leipzig 04103, Germany ; Faculty of Biological Sciences, Friedrich Schiller University Jena, Fürstengraben 1, Jena 07743, Germany
Wäldchen, Jana ; Department of Biogeochemical Integration, Max Planck Institute for Biogeochemistry, Hans Knöll Strasse 10, Jena 07745, Germany
de Haan, Myriam; Research and Collection Departments, Botanic Garden Meise, Nieuwelaan 38, Meise B-1860, Belgium
Ballings, Petra; Research and Collection Departments, Botanic Garden Meise, Nieuwelaan 38, Meise B-1860, Belgium
Van der Beeten, Iris; Research and Collection Departments, Botanic Garden Meise, Nieuwelaan 38, Meise B-1860, Belgium
Goffinet, Bernard ; Department of Ecology and Evolutionary Biology, University of Connecticut, 75 North Eagleville Road, Unit 3043, Storrs, CT 06269-3043, USA
Vanderpoorten, Alain ; Université de Liège - ULiège > Département de Biologie, Ecologie et Evolution > Biologie de l'évolution et de la conservation - Unité aCREA-Ulg (Conseils et Recherches en Ecologie Appliquée) ; Research and Collection Departments, Botanic Garden Meise, Nieuwelaan 38, Meise B-1860, Belgium
✱ These authors have contributed equally to this work.
Language :
English
Title :
Towards the automatized identification of moss species from their spore morphology.
Alfayate C, Ron E, Estébanez B, Pérez-Batista MÁ. 2013. Mature spores of four pleurocarpous mosses in the Canary Islands: ultrastructure and early germination stages. The Bryologist 116: 97-112. doi: 10.1639/0007-2745-116.2.097
Anderson TF. 1951. Techniques for the preservation of three-dimensional structure in preparing specimens for the electron microscope. Transactions of the New York Academy of Sciences 13: 130-134. doi: 10.1111/j.2164-0947.1951.tb01007.x
Aslan Z, Çulha H, Ezer T, Doǧan C. 2022. Spore morphologies of some acrocarpous mosses (Bryophyta): taxonomical and ecological significance. Anatolian Bryology 8: 106-113. doi: 10.26672/anatolianbryology.1189719
Barbé M, Fenton NJ, Bergeron Y. 2016. So close and yet so far away: long-distance dispersal events govern bryophyte metacommunity reassembly. The Journal of Ecology 104: 1707-1719. doi: 10.1111/1365-2745.12637
Barré P, Stöver BC, Müller KF, Steinhage V. 2017. LeafNet: a computer vision system for automatic plant species identification. Ecological Informatics 40: 50-56. doi: 10.1016/j.ecoinf.2017.05.005
Bechteler J, Peñaloza-Bojacá G, Bell D, et al. 2023. Comprehensive phylogenomic time tree of bryophytes reveals deep relationships and uncovers gene incongruences in the last 500 million years of diversification. American Journal of Botany 110: e16249. doi: 10.1002/ajb2.16249
Berganzo-Besga I, Orengo HA, Lumbreras F, Ramsey MN. 2025. Deep learning black box and pattern recognition analysis using Guided Grad-CAM for phytolith identification. Annals of Botany 136: 355-366. doi: 10.1093/aob/mcaf088
Boros Á. 1975. An atlas of recent European moss spores. Budapest: Akadémiai Kiadó.
Brown RC, Lemmon BE, Shimamura M, Villarreal JC, Renzaglia KS. 2015. Spores of relictual bryophytes: diverse adaptations to life on land. Review of Palaeobotany and Palynology 216: 1-17. doi: 10.1016/j.revpalbo.2015.01.004
Brubaker LB, Anderson PM, Murray BM, Koon D. 1998. A palynological investigation of true-moss (Bryidae) spores: morphology and occurrence in modern and late Quaternary lake sediments of Alaska. Canadian Journal of Botany: Journal Canadien De Botanique 76: 2145-2157. doi: 10.1139/b98-192
Cao T, Vitt D. 1986. Spore surface structure of Sphagnum. Nova Hedwigia 43: 191-220.
Carrión JS, Cano MJ, Guerra J. 1995. Spore morphology in the moss genus Pterygoneurum Jur. (Pottiaceae). Nova Hedwigia 61: 481-496. doi: 10.1127/nova.hedwigia/61/1995/481
Cronberg N, Molau U, Sonesson M. 1997. Genetic variation in the clonal bryophyte Hylocomium splendens at hierarchical geographical scales in Scandinavia. Heredity 78: 293-301. doi: 10.1038/hdy.1997.44
Dhyani A, Kasana S, Uniyal PL. 2024. From barcodes to genomes: a new era of molecular exploration in bryophyte research. Frontiers in Plant Science 15: 1500607. doi: 10.3389/fpls.2024.1500607
Dickson J. 1986. Bryophyte analysis. In: Berglund B. ed. Handbook of Holocene palaeoecology and palaeohydrology. Chichester: Wiley, 627-643.
Dyer RJ, Pellicer J, Savolainen V, Leitch IJ, Schneider H. 2013. Genome size expansion and the relationship between nuclear DNA content and spore size in the Asplenium monanthes fern complex (Aspleniaceae). BMC Plant Biology 13: 219. doi: 10.1186/1471-2229-13-219
Eppley SM, Taylor PJ, Jesson LK. 2007. Self-fertilization in mosses: a comparison of heterozygote deficiency between species with combined versus separate sexes. Heredity 98: 38-44. doi: 10.1038/sj.hdy.6800900
Estébanez B, Alfayate C, Ron E. 1997. Observations on spore ultrastructure in six species of Grimmia (Bryopsida). Grana 36: 347-357. doi: 10.1080/00173139709362628
Estébanez B, Yamaguchi T, Deguchi H. 2006. Ultrastructure of the spore in four Japanese species of Ptychomitrium Fürnr. (Musci). Grana 45: 61-70. doi: 10.1080/00173130600555722
Fröhlich-Nowoisky J, Kampf CJ, Weber B, et al. 2016. Bioaerosols in the Earth system: climate, health, and ecosystem interactions. Atmospheric Research 182: 346-376. doi: 10.1016/j.atmosres.2016.07.018
Goffinet B, Buck WR, Shaw AJ. 2008. Morphology, anatomy, and classification of the Bryophyta. In: Goffinet B, Shaw AJ. eds. Bryophyte biology, 2nd edn. Cambridge: Cambridge University Press, 55-138.
Goffinet B, Buck WR, Shaw AJ. 2009. Addenda to the classification of mosses. I. Andreaeophytina stat. nov. and Andreaeobryophytina stat. nov. The Bryologist 112: 856-857. doi: 10.1639/0007-2745-112.4.856
Goodfellow I, Bengio Y, Courville A. 2018. Deep learning. Genetic Programming and Evolvable Machines 19: 305-307.
Gunduz H, Gunal S. 2024. A lightweight convolutional neural network (CNN) model for diatom classification: DiatomNet. PeerJ: Computer Science 10: e1970. doi: 10.7717/peerj-cs.1970
Holyoak DT. 2010. Notes on taxonomy of some European species of Ephemerum (Bryopsida: Pottiaceae). Journal of Bryology 32: 122-132. doi: 10.1179/037366810X12578498136192
Horton DG. 1982. The evolutionary significance of superficial spore characters in the Bryidae. Journal of the Hattori Botanical Laboratory 53: 99-105.
Hutsemékers V, Hardy OJ, Vanderpoorten A. 2013. Does water facilitate gene flow in spore-producing plants? Insights from the fine-scale genetic structure of the aquatic moss Rhynchostegium riparioides (Brachytheciaceae). Aquatic Botany 108: 1-6. doi: 10.1016/j.aquabot.2013.02.001
Ireland R. 1987. Scanning electron microscope study of the spores of the North American species of Plagiothecium (Musci). Memoirs of the New York Botanical Garden 45: 95-110.
Katul GG, Porporato A, Nathan R, et al. 2005. Mechanistic analytical models for long-distance seed dispersal by wind. The American Naturalist 166: 368-381. doi: 10.1086/432589
Kiebacher T, Szövényi P. 2024. Morphological, genetic and ecological divergence in near-cryptic bryophyte species widespread in the Holarctic: the Dicranum acutifolium complex (Dicranales) revisited in the Alps. Journal of Plant Research 137: 561-574. doi: 10.1007/s10265-024-01534-3
Kloster M, Langenkämper D, Zurowietz M, Beszteri B, Nattkemper TW. 2020. Deep learning-based diatom taxonomy on virtual slides. Scientific Reports 10: 14416. doi: 10.1038/s41598-020-71165-w
Kučera J, Kuznetsova OI, Manukjanová A, Ignatov MS. 2019. A phylogenetic revision of the genus Hypnum: towards completion. Taxon 68: 628-660. doi: 10.1002/tax.12095
Kungu E, Longton RE, Bonner L. 2007. Character reduction and peristome morphology in Entodontaceae: constraints on an information source. In: Newton A, Tangney R. eds. Pleurocarpous mosses. Systematics and evolution. Boca Raton: CRC Press, 247-268.
Kwon DH, Lee MJ, Jeong H, Park S, Cho KH. 2025. Multi-modal learning-based algae phyla identification using image and particle modalities. Water Research 275: 123172. doi: 10.1016/j.watres.2025.123172
Lacey ME, West JS. 2006. The air spora: a manual for catching and identifying airborne biological particles. Dordrecht: Springer.
Longton RE. 1997. Reproductive biology and life-history strategies. Advances in Bryology 6: 65-101.
Lönnell N, Hylander K, Jonsson BG, Sundberg S. 2012. The fate of the missing spores - patterns of realized dispersal beyond the closest vicinity of a sporulating moss. PLoS One 7: e41987. doi: 10.1371/journal.pone.0041987
Mäder P, Boho D, Rzanny M, et al. 2021. The Flora Incognita app - interactive plant species identification. Methods in Ecology and Evolution 12: 1335-1342. doi: 10.1111/2041-210X.13611
Matavulj P, Panić M, Šikoparija B, Tešendić D, Radovanović M, Brdar S. 2023. Advanced CNN architectures for pollen classification: design and comprehensive evaluation. Applied Artificial Intelligence: AAI 37: 2157593. doi: 10.1080/08839514.2022.2157593
Matlack GR. 1987. Diaspore size, shape, and fall behavior in wind-dispersed plant species. American Journal of Botany 74: 1150-1160. doi: 10.2307/2444151
Meagher TR, Shaw J. 1990. Clonal structure in the moss, Climacium americanum Brid. Heredity 64: 233-238. doi: 10.1038/hdy.1990.28
Medina NG, Estébanez B. 2014. Does spore ultrastructure mirror different dispersal strategies in mosses? A study of seven Iberian Orthotrichum species. PLoS One 9: e112867. doi: 10.1371/journal.pone.0112867
Medina NG, Estébanez B, Francisco L, Mazimpaka V. 2009. On the presence of dimorphic spores in Orthotrichum affine (Bryopsida, Orthotrichaceae). Journal of Bryology 31: 124-126. doi: 10.1179/174328209X415086
Mikulášková E, Peterka T, Šmerda J, Hájek M. 2024. Next-generation sequencing reveals hidden genomic diversity in glacial relicts: a case study of Meesia triquetra. Journal of Systematics and Evolution 62: 475-488. doi: 10.1111/jse.13005
Mogensen GS. 1983. The spore. In: Schuster RM. ed. New manual of bryology. Nichinan, Japan: Hattori Botanical Laboratory, 325-342.
Nieto-Lugilde M, Nieto-Lugilde D, Piatkowski B, et al. 2024. Ecological differentiation and sympatry of cryptic species in the Sphagnum magellanicum complex (Bryophyta). American Journal of Botany 111: e16401. doi: 10.1002/ajb2.16401
Niklas KJ. 1985. The aerodynamics of wind pollination. The Botanical Review 51: 328-386. doi: 10.1007/BF02861079
Passarella MDA, Luizi-Ponzo AP. 2019. Palynology of Amphidium Schimp. (Amphidiaceae M. Stech): can spore morphology circumscribe the genus? Acta Botanica Brasilica 33: 135-140. doi: 10.1590/0102-33062018abb0328
Patel N, Fawcett S, Sundue M, Budke JM. 2019. Evolution of perine morphology in the Thelypteridaceae. International Journal of Plant Sciences 180: 1016-1035. doi: 10.1086/705588
Patiño J, Vanderpoorten A. 2018. Bryophyte biogeography. Critical Reviews in Plant Sciences 37: 175-209. doi: 10.1080/07352689.2018.1482444
Patiño J, Carine M, Fernández-Palacios JM, Otto R, Schaefer H, Vanderpoorten A. 2014. The anagenetic world of spore-producing land plants. New Phytologist 201: 305-311. doi: 10.1111/nph.12480
Pilkington S. 2022. Microbryum davallianum (Sm.) R.H.Zander in Britain and Ireland. Field Bryology 127: 2-7.
Punyasena SW, Haselhorst DS, Kong S, Fowlkes CC, Moreno JE. 2022. Automated identification of diverse Neotropical pollen samples using convolutional neural networks. Methods in Ecology and Evolution 13: 2049-2064. doi: 10.1111/2041-210X.13917
Renzaglia K, McFarland K, Smith D. 1997. Anatomy and ultrastructure of the sporophyte of Takakia ceratophylla (Bryophyta). American Journal of Botany 84: 1337. doi: 10.2307/2446132
Ščevková J, Štefániková N, Dušička J, Lafférsová J, Zahradníková E. 2024. Long-term pollen season trends of Fraxinus (ash), Quercus (oak) and Ambrosia artemisiifolia (ragweed) as indicators of anthropogenic climate change impact. Environmental Science and Pollution Research International 31: 43238-43248. doi: 10.1007/s11356-024-34027-w
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. 2017. Grad-CAM: visual explanations from deep networks via gradient-localization. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy, pp. 618-626. doi: 10.1109/ICCV.2017.74
Smith AJE. 2004. The moss flora of Britain and Ireland. Cambridge: Cambridge University Press.
Smith GL. 1974. New developments in the taxonomy of Polytrichaceae: epiphragm structure and spore morphology as generic characters. Journal of the Hattori Botanical Laboratory 38: 143-150.
Sundberg S. 2005. Larger capsules enhance short-range spore dispersal in Sphagnum, but what happens further away? Oikos 108: 115-124. doi: 10.1111/j.0030-1299.2005.12916.x
Sundberg S. 2013. Spore rain in relation to regional sources and beyond. Ecography 36: 364-373. doi: 10.1111/j.1600-0587.2012.07664.x
TorchVision maintainers and contributors. 2016. TorchVision: PyTorch's computer vision library. GitHub repository. https://github.com/pytorch/vision (14 April 2025, Date accessed)
Truong MXA, Van der Wal R. 2024. Exploring the landscape of automated species identification apps: development, promise, and user appraisal. BioScience 74: 601-613. doi: 10.1093/biosci/biae077
Vitt DH. 1984. Classification of the Bryospida. In: Schuster RM. ed. New manual of bryology. Nichinan: Hattori Botanical Laboratory, 696-759.
Wäldchen J, Mäder P. 2018. Plant species identification using computer vision techniques: a systematic literature review. Archives of Computational Methods in Engineering: State of the Art Reviews 25: 507-543. doi: 10.1007/s11831-016-9206-z
Wäldchen J, Rzanny M, Seeland M, Mäder P. 2018. Automated plant species identification - trends and future directions. PLOS Computational Biology 14: e1005993. doi: 10.1371/journal.pcbi.1005993
Wang Q-H, Jia Y. 2012. A taxonomic revision of the Asian species of Ulota Mohr (Orthotrichaceae). The Bryologist 115: 412-443. doi: 10.1639/0007-2745-115.3.412
Xie S, Girshick R, Dollár, P, Tu Z, He K. 2017. Aggregated residual transformations for deep neural networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, pp. 5987-5995, doi: 10.1109/CVPR.2017.634
Zanatta F, Patiño J, Lebeau F, et al. 2016. Measuring spore settling velocity for an improved assessment of dispersal rates in mosses. Annals of Botany 118: 197-206. doi: 10.1093/aob/mcw092
