This paper is the culmination of my PhD at Cornell, and I'm so grateful to the team that made it possible. Co-led with @fishyomics.bsky.social, and co-authored with @jessi-rick.bsky.social, @arynwilder.bsky.social, Zofia Baumann, David Conover, Hannes Baumann, and @ntherk.bsky.social.
Each inversion responds to distinct environmental pressures with contrasting patterns of selection, together sustaining continuous variation in multiple traits, and challenging the view of inversions as simple binary switches between ecotypes.
These inversions harbor the strongest signals of divergent selection genome-wide and underlie multiple adaptive traits. Trait associations are validated in wild and experimental populations, and genes related to growth and thermal stress inside inversions show temperature-dependent expression.
We combined QTL mapping, range-wide population genomics, transcriptomics, and lab selection experiments to link genotype, phenotype, and selection across the species' range. All lines of evidence converged on three large chromosomal inversions as key modulators of continuous local adaptation.
Our paper is out in @Science! The Atlantic silverside spans Earth's steepest latitudinal gradient in coastal sea-surface temperature. Despite high gene flow, populations show clinal genetic variation in multiple locally adapted traits. doi.org/10.1126/scie...
New review!
Theory & a practical guide to structural variants in popgenπ§¬
Many thanks to my co-authors: @rebekahoomen.bsky.social
@annatigano.bsky.social @marenwellenreuther.bsky.social @janawold.bsky.social @dlfield.bsky.social @clairemerot.bsky.social
onlinelibrary.wiley.com/doi/10.1111/...
Action Network petition: Cancel ChatGPT Edu. Invest in Humans. In February 2025, the California State University system announced a $17 million contract with OpenAI to provide ChatGPT Edu to all faculty, staff, and students on its 22 CSU campuses as part of a larger βAI-Empowered Universityβ initiative. This is the largest contract ever established between a university system and an artificial intelligence company. Despite the name, ChatGPT Edu is not educational technology. It is a general-purpose chatbot that is not designed, trained, or optimized for education. Beyond its privacy and security features, ChatGPT Edu is identical to the free online version of ChatGPT. ChatGPT Edu does not use reliable peer-reviewed sources to answer students' questions and is indifferent to whether its answers are correct. Experts argue that ChatGPT Edu is harmful to academic working conditions, diminishes the quality of teaching and learning, introduces new forms of discrimination, and is dangerous to students' mental health. Recent polling shows that CSU students share these concerns, with the majority of students expressing that they are worried about the negative impacts of generative AI on human creativity and the environment.
Cal State Universityβs deal with OpenAI β providing ChatGPT to all faculty, students, and staff β will expire in June 2026. Amid the prospect of layoffs in the CSU, weβre asking the chancellor not to renew this costly and demoralizing contract.
Link below and anyone can sign:
Green line graph time series of average sea surface temperature anomalies for each year period from 1900 through 2025. There is large interannual variability, but an overall long-term increasing trend. Anomalies are computed relative to a 1981-2010 baseline.
Last year observed the 3rd warmest global average sea surface temperature on record. π
Data from psl.noaa.gov/data/gridded...
Team fish - we need your help! We are trying to build a database of all the fish chromosome-scale genomes where sex chromosomes have been identified. Have you build one or some? Do you know someone who has? Can you post the link in the comments? Please spread the word and repost! Thank you!
Integrative PhD position available in my lab on the mechanisms (gene expression, metabolism, microbiome and hormones) of intestinal remodelling in lampreys. Come work in a great and fun research environment with @nealdawson.bsky.social & Adam Dobson (not here). Apply: www.gla.ac.uk/colleges/mvl...
Wisconsin Evolution is accepting applications for our Seminar Series' Early Career Scientist Award. Come share your evolution research and visit UW-Madison's evolution community. Open to grad students and postdocs (<5 yrs post PhD) from outside UW-Madison.
Apply by Dec 15th here: shorturl.at/4a4O6
Excited to recruit a new PhD student for Fall 2026 in my lab at Cornell! Looking for someone interested in evolutionary genomics + fisheries/conservation applications. Quick timeline this yearβreach out soon. More details: www.therkildsenlab.org/join-us.html
Thanks! Thatβs something weβre aware of but didnβt deal with in this study - itβs actually the first thing we mention in our limitations section. Itβs definitely an important consideration that warrants investigation in future work.
Now published in Cell! We found that ~15% of SNPs from divergent refs did not liftover as SNPs in the gray fox refβhalf mapped to monomorphic sites, half failed to map. Co-authored with Matthew Genchev, @elliecat.bsky.social, and @jazlynmooney.bsky.social
www.sciencedirect.com/science/arti...
Thrilled to have a new paper out where we use 12+ years of monitoring data to document the impacts of inbreeding on fitness in eastern massasauga rattlesnakes! www.pnas.org/doi/10.1073/...
Iβve been meaning to write this for a whileβ¦ but ADHD π
I finally did! Itβs about creating academic spaces that actually work for people with ADHD. If it resonates, Iβd love to connect or hear your thoughts.
www.insidehighered.com/opinion/care...
#ADHD #Neurodiversity #HigherEd
On this weekβs ep, Kevin Bird (@stairwaytokevin.bsky.social) explains the events of the last 3 months, how science is funded in the United States, why it is being attacked so aggressively, and the ideologies behind these attacks. Found everywhere you get podcasts.
open.spotify.com/episode/0E9l...
@akopyan.bsky.social et al. analysed 168 genomes of Atlantic silverside, showing that high gene flow leads to clustered divergence in large inversions, while low recombination at centromeres mimics differentiation.
π doi.org/10.1093/molb...
#evobio #molbio #inversions
Huge thanks to my incredible co-authors Anna Tigano (@annatigano.bsky.social), Arne Jacobs (@fishyomics.bsky.social), Aryn Wilder, and Nina Therkildsen for their brilliant contributions to this work. This truly represents years of collaborative effort across multiple institutions! π§¬π 7/7
This study offers valuable empirical insights distinguishing the roles of inversions (conditionally low-recombining) vs. centromeres (consistently low-recombining), illuminating the critical connection between genome structure and local adaptation with gene flow. 6/7
Meanwhile, putative centromeric regions showed high differentiation but LOW sequence divergenceβsuggesting they're unlikely to contribute to adaptive divergence with gene flow, despite their recombination-suppressing properties. 5/7
The most fascinating discovery? Different genomic features play distinct roles: Chromosomal inversions showed both high differentiation AND high sequence divergenceβevidence they maintain locally adapted alleles despite gene flow. 4/7
With increasing gene flow, we observed more clustering of differentiated regions in the genome! This supports theoretical predictions that high gene flow favors concentrated genomic architectures of adaptation. 3/7
We found a continuum of genome-wide differentiation increasing from south to north, reflecting higher connectivity among southern populations and reduced gene flow at northern latitudes. 2/7
Excited to share our new paper on genetic differentiation in Atlantic silversides! We investigated how genome structure influences adaptation under varying levels of gene flow using a species-specific reference genome and multiple recombination maps. Read on for insights! doi.org/10.1093/molb... 1/7
7/7 π¦ Huge thanks to my amazing co-authors Matthew Genchev @jazlynmooney.bsky.social @elliecat.bsky.social and shoutout to Urocyon cinereoargenteus, our beautiful study species! πΎπΈ #Teamwork #Genomics
6/7 π Why It Matters: With most species lacking conspecific references, genomic studies often rely on divergent genomes, risking biased population genomic inferences. Reference genome choice mattersβuse species-matched references or reference-free methods when possible to ensure accurate insights.
5/7 π― Selection Scans: Arctic fox and dog references identified more than twice the number of unique FST outlier windows compared to the gray fox reference. Each reference identified distinct gene ontology terms underlying outliers, reflecting different biological processes.
4/7 β³ Demography & Recombination: The gray fox genome produced higher and more stable estimates of historical population sizes. Additionally, recombination rates inferred from heterospecific genomes were inflated, particularly at chromosome ends, compared to the gray fox genome.
3/7 π Population Genomics: The conspecific gray fox genome detected more SNPs, especially rare variants, and yielded higher estimates of nucleotide diversity (Ο) and differentiation (FST) compared to dog and Arctic fox genomes.
