Biology subjects

LaFrance, R.

Publications and source records attributed to LaFrance, R..

5 recordsLinked to original sources

bioRxiv2026

Regional trait-threat interactions determine extinction risk in plants.

Global plant biodiversity and associated ecosystem services are being severely threatened by human activities and land-use change. This has led to an immense effort to assess the extinction risk of individual species and associated drivers to inform conservation actions and priorities. However, we lack a more general understanding of the processes that determine variation in vulnerability and resilience among plant taxa facing the same threats in the same or similar ecological conditions and geographic regions. This research helps identify these generalities and determines species and traits that are at disproportionate risk of extinction. Given Floridas high levels of biodiversity and endemism, coupled with ongoing rapid and extensive human development, we use the states native flora as our model system. By combining data on extinction risk, human-induced threats, and plant functional traits, our results show that species threatened with extinction cluster within trait space, with smaller herbaceous species at greater risk than larger woody species. We also find that all categories of threat considered in our study (Development and Infrastructure; Environmental Change and Modification; Natural Hazards and Climate Change; and Resource Extraction) increase extinction risk except for the Invasive Species, Pests, and Diseases threat category. Generalised linear mixed-effect models show differing responses of species to environmental change, habitat loss, and fragmentation, driven by differences in life-history strategies and dispersal capabilities and mechanisms. The varied responses of species facing the same threats uncovered here highlight that plant functional traits in the context of regional ecosystem processes can be used to predict species-specific responses to threats and land-use change, providing a valuable resource for conservation management and prioritisation. Article impact statementPlant functional traits can be used to estimate species-specific responses to human-induced threats and predict overall extinction risk.

ecology↗

bioRxiv2025

Petal to the metal: The slow road to automating large-scale phenology labeling for herbarium specimens

Herbarium specimens represent critical historical records of plant phenology, yet automating annotation of reproductive structures remains challenging given the diversity of floral morphologies, specimen age and quality, and image quality. Here, we present a machine learning pipeline that uses an ensemble modeling approach to detect flowers on herbarium specimens and deliver these data to the phenology research community. After testing multiple strategies for generating training data, we found in-house expert-curated annotations were essential for producing reliable results. Expert validation found relatively strong accuracy for detecting present floral structures, but still had moderately high false negative rates. Applying the ensemble to our filtered final image dataset of 22 million records resulted in 11.1 million records labeled with flowers present. However, only 2.9 million of these contained complete metadata necessary for downstream phenology research, highlighting the need for full label digitization efforts. Still, this dataset represents a large compilation of historical herbarium-derived phenology records available as a resource for the phenology community. We end by demonstrating how integrating these machine-labeled records into Phenobase, a publicly-available phenology database, expands taxonomic and temporal coverage for large-scale phenological analyses, and discuss remaining challenges and next steps.

ecology↗

bioRxiv2024

Ensemble automated approaches for producing high quality herbarium digital records

One of the slowest steps in digitizing natural history collections is converting labels associated with specimens into a digital data record usable for collections management and research. Recent work has shown a path for semi-automated approaches that can find labels, OCR them and convert the raw OCR text into digital data records. Here we address how raw OCR can be converted into a digital data record via extraction into standardized Darwin Core fields. We first showcase development of a rule-based approach and compare outcomes with a large language model-based approach, in particular ChatGPT4. We next quantified error rates in a set of OCRed labels, determining omission and commission errors for both approaches and documenting other issues. For example, we find that ChatGPT4 will often create field names that are not Darwin Core compliant. Our results suggest that these approaches each have different limitations. Therefore, we developed an ensemble approach that utilizes outputs from both in order to reduce problems from each individual method. An ensemble method reduces issues with field name heterogeneity and strongly reduces information extraction errors. This suggests that such an ensemble method is likely to have particular value for creating digital data records, even for complicated label content, given that longer labels, though more error prone, are still successfully extracted. While human validation is still much needed to ensure the best possible quality, we showcase working solutions to speed digitization of herbarium specimen labels that are likely usable more broadly for all natural history collection types.

bioinformatics↗

bioRxiv2022

Two shifts in evolutionary lability underlie independent gains and losses of root-nodule symbiosis in a single clade of plants

Root nodule symbiosis (RNS) allows plants to access atmospheric nitrogen converted into usable forms through a mutualistic relationship with soil bacteria. RNS is a complex trait requiring coordination from both the plant host and the bacterial symbiont, and pinpointing the evolutionary origins of root nodules is critical for understanding the genetic basis of RNS. This endeavor is complicated by data limitations and the intermittent presence of RNS in a single clade of ca. 30,000 species of flowering plants, i.e., the nitrogen-fixing clade (NFC). We developed the most extensive de novo phylogeny for all major lineages of the NFC and an enhanced root nodule trait database to reconstruct the evolution of RNS. Through identification of the evolutionary pathway to RNS gain, we show that shifts among heterogeneous evolutionary rates can explain how a complex trait such as RNS can arise many times across a large phylogeny. Our analysis identifies a two-step process in which an ancestral precursor state gave rise to a more labile state from which RNS was quickly gained at specific points in the NFC. Our rigorous reconstruction of ancestral states illustrates how a two-step pathway could have led to multiple independent gains and losses of RNS, contrary to recent hypotheses invoking just a single gain and numerous losses. RNS may be an example of multi-level convergent evolution, thus requiring a broader phylogenetic and genetic scope for genome-phenome mapping to elucidate mechanisms enabling fully functional RNS.

evolutionary biology↗

bioRxiv2019

Research applications of primary biodiversity databases in the digital age

We are in the midst of unprecedented change--climate shifts and sustained, widespread habitat degradation have led to dramatic declines in biodiversity rivaling historical extinction events. At the same time, new approaches to publishing and integrating previously disconnected data resources promise to help provide the evidence needed for more efficient and effective conservation and management. Stakeholders have invested considerable resources to contribute to online databases of species occurrences and genetic barcodes. However, estimates suggest that only 10% of biocollections are available in digital form. The biocollections community must therefore continue to promote digitization efforts, which in part requires demonstrating compelling applications of the data. Our overarching goal is therefore to determine trends in use of mobilized species occurrence data since 2010, as online systems have grown and now provide over one billion records. To do this, we characterized 501 papers that use openly accessible biodiversity databases. Our standardized tagging protocol was based on key topics of interest, including: database(s) used, taxa addressed, general uses of data, other data types linked to species occurrence data, and data quality issues addressed. We found that the most common uses of online biodiversity databases have been to estimate species distribution and richness, to outline data compilation and publication, and to assist in developing species checklists or describing new species. Only 69% of papers in our dataset addressed one or more aspects of data quality, which is low considering common errors and biases known to exist in opportunistic datasets. Globally, we find that biodiversity databases are still in the initial stages of data compilation. Novel and integrative applications are restricted to certain taxonomic groups and regions with higher numbers of quality records. Continued data digitization, publication, enhancement, and quality control efforts are necessary to make biodiversity science more efficient and relevant in our fast-changing world.

zoology↗