Reconstructing chytrid invasion and amphibian resilience using sedimentary DNA
This project is a collaboration with fellow Penn State Ecology PhD student, William Vuyk.
Collecting freshwater sediment cores from an alpine lake in Sierra National Forest. 📷 Anton Sorokin (2026)
Left panel: Inflating our kayaks with the help of Zippy, the University of Akron’s mascot, who happened to be visiting the field station while we were sampling. Right panel: A core collected from Bath Pond at the University of Akron Field Station.
The human-mediated spread of invasive species is a major threat to environmental and human health. Many species, pathogens especially, can spread globally before they are detected. This creates “detection lags” that confound our understanding of how biological invasions impact ecological communities over time. Batrachochytrium dendrobatidis (Bd), a fungal pathogen of amphibians, is now found worldwide and is linked to more extinctions than any other known infectious disease. Despite its impact, the timing, spread, and consequences of Bd’s emergence during its extended detection lag remain largely unknown.
We propose to reconstruct the cryptic invasion of Bd in North America using freshwater sediment cores. These sediments preserve continuous temporal records of DNA, which is shed into the environment by aquatic organisms and deposited in chronological layers, providing a powerful archive of amphibian-Bd interactions through time. Using qPCR, Bd target capture sequencing, and amphibian metabarcoding, we aim to detect historical Bd presence and host amphibian community composition changes in the sedimentary archive.
We have collected cores from sites in California’s Sierra Nevada, where Bd arrival and subsequent frog declines were well documented, and sites across Pennsylvania and Ohio’s Allegheny Plateau, where Bd is ubiquitous but there are no documented amphibian declines. Sampling a region with known decline history (the Sierra Nevada) will allow us to validate our historical inferences before applying our methods to a region with unknown history (the Allegheny Plateau). Ultimately, we aim to better reconstruct the history of Bd invasion across North America, and how that history is influences modern patterns of Bd susceptibility.
Left panel: A sediment core pulled from a lake in Sierra National Forest. Middle panel: A core collected in Yosemite National Park with visible annually layered sediments. Right panel: Preparing to core in Yosemite National Park.
A critically endangered Rana Sierrae basks in the foreground while we prepare our inflatable kayaks for sediment core collection in Sierra National Forest. 📷 Anton Sorokin (2026)
Linking cryptic chytrid invasion to historically enigmatic amphibian declines in Brazil
Left panel: Sampling a large Haddadus binotatus at the Smithsonian’s National Museum of Natural History (NMNH). Middle panel: Pulling jars from the collection at the Smithsonian NMNH’s Division of Amphibians and Reptiles. Right panel: Historical DNA extraction at the Smithsonian’s Center for Conservation Genomics’ Ancient DNA Lab.
As human activity accelerates the spread of pathogens and other invasive species, reconstructing the cryptic early dynamics of undetected invasions has become a central challenge in disease ecology and invasion biology. In this project, my collaborators and I used museum specimens to reconstruct cryptic invasion of Batrachochytrium dendrobatidis (Bd), a fungal panzootic driving global amphibian biodiversity loss, across historically enigmatic declines in the Brazilian Atlantic Forest.
Two hypotheses have been proposed to explain Bd’s historical dynamics in this region: long-term enzootic persistence versus recent epizootic invasion coinciding with declines. Using molecular screening, histopathology, and genotyping of museum and modern specimens, we found no evidence of Bd infection prior to the decline, followed by widespread infection immediately thereafter. Only the global panzootic lineage (Bd-GPL) was detected during the initial post-decline period, implicating its invasion in driving regional losses. Infection prevalence varied among species following the decline, likely reflecting species-specific interactions between baseline susceptibility and ecological exposure risk. Contemporary host populations exhibit more uniform prevalence regardless of decline history, consistent with the evolution of pathogen tolerance under enzootic conditions.
Together, these results provide the first site-specific evidence linking Bd-GPL invasion to enigmatic amphibian declines in Brazil and demonstrate the power of museum collections for resolving detection lag dynamics in emerging infectious diseases.
Bd infection prevalence at the Boracéia Biological Station peaked following the enigmatic 1979 amphibian declines and later stabilized. Black diamonds represent observed annual Bd infection prevalence as estimated by qPCR with 95% binomial confidence intervals. The red line shows GAM-predicted prevalence through time with 95% confidence intervals (red ribbon).
Evolution of non-visual opsin genes across ecological transitions in frogs
Our paper was highlighted in press releases by Penn State and York University!
Examining our frog phylogeny (Photo: James Di Loreto, Smithsonian Institution)
Nonvisual opsins are transmembrane proteins expressed in the eyes and other tissues of many animals. When paired with a light-sensitive chromophore, nonvisual opsins form photopigments involved in various nonvisual, light-detection functions including circadian rhythm regulation, light-seeking behaviors, and seasonal responses.
In this project, we investigated the molecular evolution of nonvisual opsin genes in frogs. We tested several evolutionary hypotheses including the predicted loss of nonvisual opsins due to nocturnal ancestry and potential functional differences in nonvisual opsins resulting from environmental light variation across diverse frog ecologies. Using whole-eye transcriptomes of 81 species, combined with genomes, multitissue transcriptomes, and independently annotated genes from an additional 21 species, we identified which nonvisual opsins are present in anuran genomes and those that are also expressed in the eyes, compared selective constraint among genes, and tested for potential adaptive evolution by comparing selection between discrete ecological classes.
At the genomic level, we recovered all 18 ancestral vertebrate nonvisual opsins, indicating that anurans demonstrate the lowest documented amount of opsin gene loss among ancestrally nocturnal tetrapods. We consistently found expression of 14 nonvisual opsins in anuran eyes and detected positive selection in a subset of these genes. We also found shifts in selective constraint acting on nonvisual opsins in frogs with differing activity periods, habitats, distributions, life histories, and pupil shapes, which may reflect functional adaptation. Although many nonvisual opsins remain poorly understood, these findings provide insight into the diversity and evolution of these genes across anurans, filling an important gap in our understanding of vertebrate opsins and setting the stage for future research on their functional evolution across taxa.
After many years of work, this project was published June 2024 in the journal Molecular Biology and Evolution!
