Eryn McFarlane

I'm trying a new "see how well AI image gen is" test. This time I'm going to ask it to draw a map of the world, and then zoom in to Europe, the UK, Yorkshire and York to see when things start to get hazy.

All of these are with the free models using the smart/thinking variants. 1/n

Yes, clever! I do worry that not everyone has this level of self reflection, I'm not even sure that I do all the time if I look at something and tell myself 'sure, makes sense'.

I'm so back and forth on this. I think it really does depend on the student's level of subject matter expertise. If they can tell that AI is giving them nonsense [writing or code], then they might be okay. If they can't tell, because they're supposed to be answering themselves, then it's problematic

This is so cool!

🧪

Anthropogenic factors explain wing colour variation in Colias (Lepidoptera: Pieridae) butterflies across the Toronto urbanization gradient Hybridization, or interbreeding between two previously diverged populations, is increasing due to human influences on the environment. Rates of hybridization might be increasing particularly quickly a...

Amanda Sabatino, @jpjahner.bsky.social and I have a new paper on @biorxiv-evobio.bsky.social. Amanda, in her undergraduate thesis work, found variation in Colias colour phenotypes is associated with different metrics of urbanization in Toronto. www.biorxiv.org/content/10.1...

We’re scientists who study sex and gender. Here’s what Trump gets wrong about trans people OPINION: Despite what President Trump would have you believe, transgender people are a natural part of the human species, Joan Roughgarden and Jaimie Veale write.

For so long, they said gays & lesbians were freaks who must be suppressed & converted. They lost that fight.

Now they’re saying the same about trans people.

Like gays & lesbians, trans people have been in all societies forever. They’re part of the natural order. Deal with it.

This is horrific, I'm so sorry. Congrats on your DPhil, Dr. Turner.

I'm following this for sure.

Colour-ringed long-tailed tit

🚨 Come join the fantastic long-tailed tit team 🚨
🐦We are recruiting a field technician to collect behavioural data on a long-term studied population of wild long-tailed tits (passerine birds).
🪶20 months contract in Sheffield, UK.
🌲Deadline: 23/11/2025.
jobsite.sheffield.ac.uk/job/Technici...

Postdoctoral researcher in molecular ecology - Uppsala University Postdoctoral researcher in molecular ecology, Department of Ecology and Genetics, Uppsala University

📢 We have an open position for a postdoc to join my lab. It's a great position @animecol-uu.bsky.social, fully salaried for 2.5 years with all benefits.

www.uu.se/en/about-uu/...

The project is about transmission patterns of bacteria and antimicrobial resistance (AMR) in aquatic insects. 🧬🦠 (1/3)

ooh, thank you! looks cool!

Hybridization and speciation* Abstract. Hybridization has many and varied impacts on the process of speciation. Hybridization may slow or reverse differentiation by allowing gene flow a

This is a great, if dense, overall primer on hybridization and speciation academic.oup.com/jeb/article/...

🧪

Yelling at the TV about random, useless, absolutely spurious stats is one of my great joys in watching baseball. Isn't this why we watch?

I get that the news cycle is packed right now, but I just heard from a colleague at the Smithsonian that this is fully a GIANT SQUID BEING EATEN BY A SPERM WHALE and it’s possibly the first ever confirmed video according to a friend at NOAA

10 YEAR OLD ME IS LOSING HER MIND (a thread 🧵)

I love a magic trait! I feel like @qvarnstromlab.bsky.social instilled the joy of these in me.

Right? It's delightful to be surprised like this.

This sounds so weird! I love it.

🧪 postdoc in Toronto in a very cool lab!

This infographic describes the main results from a survey of members of 14 scientific societies. There is a pie chart showing that both academic (58%) and non-academic (42%) responded to the survey. It lists statistics: 85% of respondents reported uncertainty about the future; 71% reported federal training programs were "very" or "extremely important"; 83% reported negative impacts or irreparable harm on their field of science. There is a word cloud summarizing open-ended responses with the largest words "student research fund federal grant". Eight major themes in the responses are listed: disruption of research with societal important; concerns of closure of the USGS Bird Banding Lab; Early Career Bottleneck; Restricted Freedoms including travel and speech; Decline in Government Efficiency and Expertise; Concerns about Biased/Removed Data and Data Gaps; Concerns about the Ability to Meet Legal Mandates; and From Positive Impact to Irreparable Harm. There are icons for each participating society and a QR code for website and data availability.

Today 14 scientific societies in #ecology #evolution and #marineScience are publishing the results of a survey that finds negative impacts of federal policies on food security, flood mitigation, infectious disease preparedness, and wildlife conservation: 🧪👩‍🔬 www.firsthandaccounts.org/impacts/2025...

I love city fieldwork. 17 colias samples today and back at my best for a late lunch and a student’s thesis to edit. #fieldwork #humblebrag

I've immediately passed this on to my lab and my co-instructors. Thank you for this.

Editorial: Accountability, voice, and trust - responsible use of GenAI in scientific publishing Over the last few years, the development and rapid advancement of Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) such as ChatG

Many of us are using LLMs in coding and scientific writing. They are very helpful tools, but we need to be thoughtful as we integrate them into our work flows to preserve accountability, voice and trust. academic.oup.com/evlett/advan.... An editorial at @evolletters.bsky.social for discussion:

@hybridzones.bsky.social and @jenncoughlan.bsky.social - this looks super cool! Definitely on the to-read list.

I need to say this loudly for those in the back. Shirley Chen was an undergraduate student in my lab. She is an incredible scholar, and so are many of the undergraduates that I get to interact with. Undergraduates are doing incredible research at @york-university.bsky.social.

We suggest a careful use of structural equation models to account for many mechanisms to test the relationship between the latent variables (diversity and disease) that we're really interested in.

Additionally, because there are so many component mechanisms involved in a net diversity-disease relationship, cancelling out is often considered more likely than a lack of relationship.

We found that, as they're usually stated, amplification and dilution are not falsifiable hypotheses. Often, when they're tested, there are ad hoc explanations that are brought in to protect these hypotheses. Null hypotheses are not taken as support for a lack of relationship.

Reappraisal of the Dilution and Amplification Effect Framework: A Case Study in Lyme Disease The biodiversity–disease relationship posits two hypotheses: the dilution effect (where there's a negative relationship between biodiversity and disease risk) and the amplification effect (where ther...

Shirley Chen's paper from my lab on the dilution and amplification hypotheses regarding the biodiversity-disease relationship.
onlinelibrary.wiley.com/doi/10.1002/...