Accurate identification of fungi species presents a unique challenge in computer vision due to fine-grained inter-species variation and high intra-species variation. This paper presents our approach for the FungiCLEF 2025 competition, which focuses on few-shot fine-grained visual categorization (FGVC) using the FungiTastic Few-Shot dataset. Our team (DS@GT) experimented with multiple vision transformer models, data augmentation, weighted sampling, and incorporating textual information. We also explored generative AI models for zero-shot classification using structured prompting but found them to significantly underperform relative to vision-based models. Our final model outperformed both competition baselines and highlighted the effectiveness of domain specific pretraining and balanced sampling strategies. Our approach ranked 35/74 on the private test set in post-completion evaluation, this suggests additional work can be done on metadata selection and domain-adapted multi-modal learning. Our code is available at https://github.com/dsgt-arc/fungiclef-2025.
Retrieval-Augmented Generation (RAG) represents a transformative approach within natural language processing (NLP), combining neural information retrieval with generative language modeling to enhance both contextual accuracy and factual reliability of responses. Unlike conventional Large Language Models (LLMs), which are constrained by static training corpora, RAG-powered systems dynamically integrate domain-specific external knowledge sources, thereby overcoming temporal and disciplinary limitations. In this study, we present the design and evaluation of a RAG-enabled system tailored for Mycophyto, with a focus on advancing agricultural applications related to arbuscular mycorrhizal fungi (AMF). These fungi play a critical role in sustainable agriculture by enhancing nutrient acquisition, improving plant resilience under abiotic and biotic stresses, and contributing to soil health. Our system operationalizes a dual-layered strategy: (i) semantic retrieval and augmentation of domain-specific content from agronomy and biotechnology corpora using vector embeddings, and (ii) structured data extraction to capture predefined experimental metadata such as inoculation methods, spore densi
The FungiCLEF 2025 competition addresses the challenge of automatic fungal species recognition using realistic, field-collected observational data. Accurate identification tools support both mycologists and citizen scientists, greatly enhancing large-scale biodiversity monitoring. Effective recognition systems in this context must handle highly imbalanced class distributions and provide reliable performance even when very few training samples are available for many species, especially rare and under-documented taxa that are often missing from standard training sets. According to competition organizers, about 20\% of all verified fungi observations, representing nearly 20,000 instances, are associated with these rarely recorded species. To tackle this challenge, we propose a robust deep learning method based on prototypical networks, which enhances prototype representations for few-shot fungal classification. Our prototypical network approach exceeds the competition baseline by more than 30 percentage points in Recall@5 on both the public (PB) and private (PR) leaderboards. This demonstrates strong potential for accurately identifying both common and rare fungal species, supporting
We develop a rigorous, equation-free category-theoretic foundation for fungal organisation. A fungal organism is formalised as a functor from a category $\Env$ of structured environmental states and admissible transformations to a category $\Myc$ of mycelial network states and biologically meaningful morphisms. An operational program category $\Prog$ models time-ordered exposure protocols, and a semantics functor $\mathcal{F}_{\mathrm{prog}}:\Prog\to\Myc$ maps experimental perturbations to induced network transformations. Species and strain variability are expressed as natural transformations between fungal functors, and ecological feedback is captured via an adjunction between sensing and environment modification. Network fusion (anastomosis) is identified with pushouts in $\Myc$, and order effects in exposure sequences are quantified by a local Lie structure and a Baker--Campbell--Hausdorff expansion near the identity program. A minimal worked exposure example demonstrates how non-commutativity yields experimentally testable quadratic scaling of order asymmetry. The framework provides a structurally explicit and falsifiable basis for analysing compositional perturbations, mixture
Fungi cells are capable of sensing extracellular cues through reception, transduction and response systems which allow them to communicate with their host and adapt to their environment. They display effective regulatory protein expressions which enhance and regulate their response and adaptation to a variety of triggers such as stress, hormones, light, chemicals and host factors. In our recent studies, we have shown that $Pleurotus$ oyster fungi generate electrical potential impulses in the form of spike events as a result of their exposure to environmental, mechanical and chemical triggers, demonstrating that it is possible to discern the nature of stimuli from the fungi electrical responses. Harnessing the power of fungi sensing and intelligent capabilities, we explored the communication protocols of fungi as reporters of human chemical secretions such as hormones, addressing the question if fungi can sense human signals. We exposed $Pleurotus$ oyster fungi to cortisol, directly applied to a surface of a hemp shavings substrate colonised by fungi, and recorded the electrical activity of fungi. The response of fungi to cortisol was also supplementary studied through the applicati
Fungi undergo dynamic morphological transformations throughout their lifecycle, forming intricate networks as they transition from spores to mature mycelium structures. To support the study of these time-dependent processes, we present a synthetic, time-aligned image dataset that models key stages of fungal growth. This dataset systematically captures phenomena such as spore size reduction, branching dynamics, and the emergence of complex mycelium networks. The controlled generation process ensures temporal consistency, scalability, and structural alignment, addressing the limitations of real-world fungal datasets. Optimized for deep learning (DL) applications, this dataset facilitates the development of models for classifying growth stages, predicting fungal development, and analyzing morphological patterns over time. With applications spanning agriculture, medicine, and industrial mycology, this resource provides a robust foundation for automating fungal analysis, enhancing disease monitoring, and advancing fungal biology research through artificial intelligence.
FungiCLEF 2024 addresses the fine-grained visual categorization (FGVC) of fungi species, with a focus on identifying poisonous species. This task is challenging due to the size and class imbalance of the dataset, subtle inter-class variations, and significant intra-class variability amongst samples. In this paper, we document our approach in tackling this challenge through the use of ensemble classifier heads on pre-computed image embeddings. Our team (DS@GT) demonstrate that state-of-the-art self-supervised vision models can be utilized as robust feature extractors for downstream application of computer vision tasks without the need for task-specific fine-tuning on the vision backbone. Our approach achieved the best Track 3 score (0.345), accuracy (78.4%) and macro-F1 (0.577) on the private test set in post competition evaluation. Our code is available at https://github.com/dsgt-kaggle-clef/fungiclef-2024.
The impact of droplets on concave surfaces is poorly understood, although it is relevant to a mode of dispersal that has evolved independently in several species of plants and fungi. This mode relies on splash-cups, specialized organs that use raindrops to disperse reproductive units away from the parent organism. We investigated the impact of droplets on conical cavities that mimic splash-cups and we found that such impact may lead to the formation of two types of jets, which appear essential for dispersal in nature. We built a minimal kinematic model that explains jet formation, involving the motion of fluid particles along geodesics (shortest paths) on the cone surface and we predicted cone angles that optimize jet formation, consistent with the geometries of natural splash-cups
Biological entities in nature have developed sophisticated communication methods over millennia to facilitate cooperation. Among these entities, plants are some of the most intricate communicators. They interact with each other through various communication modalities, creating networks that enable the exchange of information and nutrients. In this paper, we explore this collective behavior and its components. We then introduce the concept of agent plants, outlining their architecture and detailing the tasks of each unit. Additionally, we investigate the mycorrhizal fungi-plant symbiosis to extract glucose for energy harvesting. We propose an architecture that converts the chemical energy stored in these glucose molecules into electrical energy. We conduct comprehensive analyses of the proposed architecture to validate its effectiveness.
Incremental computations attempt to exploit input similarities over time, reusing work that is unaffected by input changes. To maximize this reuse in a general-purpose programming setting, programmers need a mechanism to identify dynamic allocations (of data and subcomputations) that correspond over time. We present Fungi, a typed functional language for incremental computation with names. Unlike prior general-purpose languages for incremental computing, Fungi's notion of names is formal, general, and statically verifiable. Fungi's type-and-effect system permits the programmer to encode (program-specific) local invariants about names, and to use these invariants to establish global uniqueness for their composed programs, the property of using names correctly. We prove that well-typed Fungi programs respect global uniqueness. We derive a bidirectional version of the type and effect system, and we have implemented a prototype of Fungi in Rust. We apply Fungi to a library of incremental collections, showing that it is expressive in practice.
We develop an unstructured mathematical model describing the growth kinetics of the Trichoderma fungus and the production of enzymes (cellulase) by degradation of a substrate (cellulose) in the rhizosphere. We integrate into this model the hydrolysis step of the organic matter and analyze the asymptotic behaviour of the obtained system. We show that our system evolves towards a global attractor consisting of infinite non-hyperbolic equilibria.
ungi exhibit oscillations of extracellular electrical potential recorded via differential electrodes inserted into a substrate colonised by mycelium or directly into sporocarps. We analysed electrical activity of ghost fungi (\emph{Omphalotus nidiformis}), Enoki fungi (\emph{Flammulina velutipes}), split gill fungi (\emph{Schizophyllum commune}) and caterpillar fungi (\emph{Cordyceps militari}). The spiking characteristics are species specific: a spike duration varies from one to 21 hours and an amplitude from 0.03~mV to 2.1mV. We found that spikes are often clustered into trains. Assuming that spikes of electrical activity are used by fungi to communicate and process information in mycelium networks, we group spikes into words and provide a linguistic and information complexity analysis of the fungal spiking activity. We demonstrate that distributions of fungal word lengths match that of human languages. We also construct algorithmic and Liz-Zempel complexity hierarchies of fungal sentences and show that species \emph{S. commune} generate most complex sentences.
Endophytes are tiny organisms present in living tissues of distinct plants and have been extensively studied for their endophytic microbial complement. Roots of Rosmarinus officinalis were subjected to the isolation of endophytic fungi and screened for antimicrobial activity against Gram-positive (Staphylococcus aureus and Bacillus subtilis) and Gram-negative (Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae) bacteria. Genomic DNA from active fungal strain of Trichoderma harzianum was isolated, and the internal transcribed spacer (ITS) region was amplified using ITS4 and ITS5 primers and sequenced for genetic inference in fungus. The crude extract of T. harzianum isolate with Ethyl acetate was showed significant antimicrobial activity against P. aeruginosa, S. aureus, K. pneumonia, B. subtilis and E. coli. The antimicrobial activity was highest against P. aeruginosa at concentration of 40 microgram/ ml, followed by S. aureus and K. pneumonia at the same concentration. The lowest antimicrobial activity was against by S. aureus at concentration of 60 microgram/ ml. The current study is confirmed that the antimicrobial activity is due to bioactive compounds founded in e
Living fungal mycelium networks are proven to have properties of memristors, capacitors and various sensors. To further progress our designs in fungal electronics we need to evaluate how electrical signals can be propagated through mycelium networks. We investigate the ability of mycelium-bound composites to convey electrical signals, thereby enabling the transmission of frequency-modulated information through mycelium networks. Mycelia were found to reliably transfer signals with a recoverable frequency comparable to the input, in the \SIrange{100}{10000} {\hertz} frequency range. Mycelial adaptive responses, such as tissue repair, may result in fragile connections, however. While the mean amplitude of output signals was not reproducible among replicate experiments exposed to the same input frequency, the variance across groups was highly consistent. Our work is supported by NARX modelling through which an approximate transfer function was derived. These findings advance the state of the art of using mycelium-bound composites in analogue electronics and unconventional computing.
Modern security, infrastructure, and safety-critical systems increasingly operate in environments characterised by disruption, uncertainty, physical damage, and degraded communications. Conventional digital technologies -- centralised sensors, software-defined control, and energy-intensive monitoring -- often struggle under such conditions. We propose fungi, and in particular living mycelial networks, as a novel class of biohybride systems for security, resilience, and protection in extreme environments. We discuss how fungi can function as distributed sensing substrates, self-healing materials, and low-observability anomaly-detection layers. We map fungal properties -- such as decentralised control, embodied memory, and autonomous repair -- to applications in infrastructure protection, environmental monitoring, tamper evidence, and long-duration resilience.
The effectiveness of zero-shot classification in large vision-language models (VLMs), such as Contrastive Language-Image Pre-training (CLIP), depends on access to extensive, well-aligned text-image datasets. In this work, we introduce two complementary data sources, one generated by large language models (LLMs) to describe the stages of fungal growth and another comprising a diverse set of synthetic fungi images. These datasets are designed to enhance CLIPs zero-shot classification capabilities for fungi-related tasks. To ensure effective alignment between text and image data, we project them into CLIPs shared representation space, focusing on different fungal growth stages. We generate text using LLaMA3.2 to bridge modality gaps and synthetically create fungi images. Furthermore, we investigate knowledge transfer by comparing text outputs from different LLM techniques to refine classification across growth stages.
As the next step in extraterrestrial exploration, many engineers and scientists around the country revealed their intense interest to enable multiplanetary human life, including colonizing Mars. This study proposes that architecture on Mars can be realized by a synthetic lichen system, composed of diazotrophic cyanobacteria and filamentous fungi, which produce abundant biominerals and biopolymers to bond Martian regolith into consolidated building blocks. These self-growing building blocks can be assembled into a wide range of structures. Diazotrophic cyanobacteria will 1) fix carbon dioxide and dinitrogen from the atmosphere and convert them into oxygen and organic carbon and nitrogen sources to support filamentous fungi; and 2) give rise to high concentrations of carbonate ions because of photosynthetic activities. Filamentous fungi will 1) bind metal ions onto fungal cell walls and serve as nucleation sites to promote biomineral precipitates; and 2) assist the survival and growth of cyanobacteria by providing them water, minerals, additional carbon dioxide, and protection. This report presents the major progress of the project. It has been tested and confirmed that such co-cultu
Sepsis is a life-threatening condition which requires rapid diagnosis and treatment. Traditional microbiological methods are time-consuming and expensive. In response to these challenges, deep learning algorithms were developed to identify 14 bacteria species and 3 yeast-like fungi from microscopic images of Gram-stained smears of positive blood samples from sepsis patients. A total of 16,637 Gram-stained microscopic images were used in the study. The analysis used the Cellpose 3 model for segmentation and Attention-based Deep Multiple Instance Learning for classification. Our model achieved an accuracy of 77.15% for bacteria and 71.39% for fungi, with ROC AUC of 0.97 and 0.88, respectively. The highest values, reaching up to 96.2%, were obtained for Cutibacterium acnes, Enterococcus faecium, Stenotrophomonas maltophilia and Nakaseomyces glabratus. Classification difficulties were observed in closely related species, such as Staphylococcus hominis and Staphylococcus haemolyticus, due to morphological similarity, and within Candida albicans due to high morphotic diversity. The study confirms the potential of our model for microbial classification, but it also indicates the need for
We introduce BioTrove, the largest publicly accessible dataset designed to advance AI applications in biodiversity. Curated from the iNaturalist platform and vetted to include only research-grade data, BioTrove contains 161.9 million images, offering unprecedented scale and diversity from three primary kingdoms: Animalia ("animals"), Fungi ("fungi"), and Plantae ("plants"), spanning approximately 366.6K species. Each image is annotated with scientific names, taxonomic hierarchies, and common names, providing rich metadata to support accurate AI model development across diverse species and ecosystems. We demonstrate the value of BioTrove by releasing a suite of CLIP models trained using a subset of 40 million captioned images, known as BioTrove-Train. This subset focuses on seven categories within the dataset that are underrepresented in standard image recognition models, selected for their critical role in biodiversity and agriculture: Aves ("birds"), Arachnida ("spiders/ticks/mites"), Insecta ("insects"), Plantae ("plants"), Fungi ("fungi"), Mollusca ("snails"), and Reptilia ("snakes/lizards"). To support rigorous assessment, we introduce several new benchmarks and report model ac
Comparing specific types of organisms as they are found across environmental conditions has helped inform how genes and gene products of these organisms relate to phenotypes and adaptation. In this study, we examine metatranscriptomic data as found for oceanic fungi across different oceanic sampling sites. A specific set of three genes was chosen for evaluation based on conserved orthology, known association with core physiological processes in fungi, and level of abundance within oceanic metatranscriptomic data. We report upon a potential association of genetic variance with environmental conditions of iron, salt and phosphate in oceanic waters based on heatmap visualization and PERMANOVA analysis.