Marta Shocket 🏳️‍🌈

Yeah sex is cool, but have you tried facultatively switching back and forth between sex and parthenogenesis to take advantage of genetic recombination in stressful situations while not wasting your reproductive output on dead-end males when conditions are good? #daphnia

Automatic variable selection is a powerful technique for simplifying models, reducing overfitting, and improving interpretability. It enables you to efficiently identify the most important predictors from a large set of variables. However, it’s important to recognize that automatic variable selection involves random processes. The same algorithm may select different variables across multiple runs, and different algorithms can yield entirely different selections, even when applied to the same data set. The graph below illustrates how different automatic variable selection methods perform across 200 simulation runs. Each box represents one method, with rows corresponding to variables and columns to simulation runs. Black indicates selected variables, while white indicates excluded ones. 🔹 Stepwise Selection: Highly inconsistent, with patterns that often appear random. Very few variables are consistently selected across all runs. 🔹 Regression Tree: Most variables are rarely selected, with only a small subset chosen consistently across simulations. The small median model size reflects this focused selection. 🔹 Random Forest: Demonstrates improved stability compared to regression trees, with more consistently selected variables, though variability persists for weaker predictors. This method tends to include a broader set of variables compared to regression trees. 🔹 Lasso and Elastic Net: Both methods exhibit relatively stable variable selection, with Elastic Net slightly outperforming Lasso due to its larger model size, allowing for broader inclusion of important variables. While no method achieves perfect consistency, Random Forest, Lasso, and Elastic Net generally provide more stable and reliable variable selection results, whereas Stepwise Selection tends to be the least reliable. I published these results in a working paper back in 2018, but they remain highly relevant today. If you're interested, you can read the full paper here: statistiques.public.lu/dam-assets/cat…. If you enjoy insights like these, subscribe to my free email newsletter for regular tips on data science, statistics, Python, and R programming. Further details: statisticsglobe.com/newsletter #datasciencetraining #RStats #VisualAnalytics #datastructure #RStudio

There are no words! Think twice about working for University of California. UCLA's Most Decorated Ecologist Just Got Fired : Here's the sequence linkedin.com/pulse/uclas-mo…

CLOSING SOON: Postdoc (University Park, USA) NIH-NSF funded project on human health, movement, and infectious diseases. iddjobs.org/jobs/the-bhart…

🚨Postdoc in Macroevolution🚨 Deadline April 12th Join a team of Stuttgart botanists & myself to test drivers of Angiosperm diversification! Submit cover letter, CV, 3 referees & publication list in a single PDF to postdoc-botany@smns-bw.de + 3 key papers (if possible) as PDF

With all eyes focused on Iran, Trump and DHS are pushing forward on border wall construction through Big Bend National Park. They will hand the Rio Grande to Mexico, destroy thousands of jobs, and ruin the last remaining wild West, all for less than 1% of all border crossings:

Combining blood meal analysis and parasite detection yields a more comprehensive understanding of insect host feeding patterns dlvr.it/TM27qS Parasites&Vectors

@docmilanfar 'calculus for babies' introduced TLS first, which is not the typical sequence!

Most tech people know about ordinary least-squares (OLS), but not so much about *total* least-squares (TLS) They measure fit differently: OLS minimizes sum of sq. distances along dependent variable; whereas TLS minimizes sum of orthogonal distances from data to fit line 1/2

PhD position (Liverpool, UK) Developing models to help make decisions early in a major disease outbreak when availability of new tests is limited. with @emjnixon at University of Liverpool More details: iddjobs.org/jobs/developin…

maybe my favorite paper I've written, I have a synthesis paper out today early access in The American Naturalist today that attempts to answer a simple but slippery question: what is an elevational range?

Looking for a job in aquatic science? Check out our job board! You don’t have to be an ASLO member to browse opportunities, from research positions to to post-docs and even industry roles. Explore now and take the next step in your career! aslo.users.membersuite.com/community/care…

House Republicans have added a provision to their sweeping tax cut package that would authorize the sale of thousands of acres of public lands in Nevada and Utah, prompting outrage from Democrats and environmental groups who called the plan a betrayal that could lead to increased drilling, mining and logging in the West. to.pbs.org/4m5xBXD

Another letter of alarm against the destructive desmantling of science in the US by the Trump administration. Now signed by over 3000 scientists to save NOAA. NOAA provides invaluable data on Earth observations and science on the liveability on our planet. ucs.org/about/news/ove…

One week left to apply! Please use the University of Glasgow portal to suit your applications! (Link on the right side of findaphd post)

The NIH plans to defund the WHI, the largest clinical trial ever for women that was meant to catch up on missed research. This study has been a fountain of knowledge. The hatred of women from this administration is really from cradle to grave. science.org/content/articl…

Global Parasitology Seminar: Sebastian Lourido on Toxoplasma biology Monday 4/21 12 pm EDT. ⁦@PennMicro⁩ ⁦@BSPparasitology⁩ ⁦@TrendsParasitol⁩ ⁦@SFParasitologie⁩ ⁦@AS_Para⁩ ⁦@AmSocParasit⁩ ⁦@parasitesrule⁩ ⁦@MBLBoP⁩ ⁦⁦⁦

Much the most off-putting thing about learning grasses is that you need to learn a completely new language. The best plan is to go out and collect a flowering Poa annua from the pavement, then try out these important new words on it.

Our new preprint (ecoevorxiv.org/repository/vie…) describes✨AedesTraits✨ (zenodo.org/records/151499…): A global database on temperature-trait responses in Aedes mosquitoes. #AedesTraits holds >31,000 rows of fully open data extracted from published literature (1930s to 2025)!

Great job leading @MartaShocket! Exciting for predictions of Malaria & general thinking abt handling fluctuating temperatures when matching theory & data. Great work @JoeyBernhardt, Courtney Murdock, Sadie Ryan, & other coauthors no longer on X nature.com/articles/s4146…