AI optimization https://brianomeara.info/posts/ai_optimization/
Brian O'Meara
I study diversification, species delimitation, trait evolution & other macroevolutionary questions, often using new methods. Also random coding projects in my own time. I'm a college professor.
Views my own. Living in Oak Ridge, TN, USA.
I often follow interesting mastodon servers by following all their users (basically, merging several local feeds). Bonus is this helps new people get at least one follower (me) quickly.
He/him
PhyloPapers 2025, Reticulate evolution https://brianomeara.info/posts/phylopapers_2025_Oct_10/
PhyloPapers 2025, Growth of the tree of life https://brianomeara.info/posts/phylopapers_2025_Sep_05/
Course ads on R-sig-phylo https://brianomeara.info/posts/rsig/
I'm taking over for the wonderful Nina Fefferman as interim director of NIMBioS, the National Institute for Modeling Biological Systems. For 17 years NIMBioS has been at the forefront of mathematics-biology, modeling, team science, and more. Much of my work wouldn't have been possible without it, and that's true for hundreds of people.
Please reach out to me with ideas for its next phase!
Funding for tenure https://brianomeara.info/posts/fundingfortenure/
Uncertainty and risk https://brianomeara.info/posts/uncertaintyandrisk/
Website and free webinar on applying to grad school in ecology and evolution, with a focus on the "#HiddenCurriculum" students might not know. Please share!
* How funding works
* How to choose advisor
* Mental health
* PhD vs Masters
* And more
Website: http://applyingtoeeb.info
Webinar: Oct 24, 11 am US Eastern, 8 am US Pacific, 15:00 UTC
Please let me know suggestions, too
We are recruiting new EvoAllies for the Evolution meetings as part of our SafeEvolution initiative. The goal of the program is to improve the climate at the meetings, making them more welcoming for everyone. It is not enforcement (we have a safety officer to investigate and sanctioning committee), but rather to help de-escalate situations, direct people to resources to help them, and to identify any problems.
Apply by June 7 at https://forms.gle/vwmupy6tuXqLijWUA
New #OpenAccess paper on evolution of root-nodule symbiosis (key in nitrogen fixation), including rate heterogeneity and both gains and losses. https://www.nature.com/articles/s41467-024-48036-3 . Team effort by multiple research groups; important for the cool #evolution and #ecology but could also have implications for agriculture (understanding how nitrogen fixation evolves and is lost)
New paper! "dentist: Quantifying uncertainty by sampling points around maximum likelihood estimates". Easy thing to plug into R workflows for getting better confidence intervals and detecting potential identifiability issues. #OpenAccess paper at https://doi.org/10.1111/2041-210X.14297, website at https://bomeara.github.io/dentist/, blog post at https://brianomeara.info/posts/dentist/
Dentist for likelihood https://brianomeara.info/posts/dentist/
Did demo for intro bio grad students on #Rstats, focused on what they could do with it, ways to organize code, etc. (they all had a little experience already).
Demo page with quarto: https://bomeara.github.io/geewhizR/
Source: https://github.com/bomeara/geewhizR/
I had them use Visual Studio (rather than closed source Rstudio) but windows users had issues with R install
@andybaio's post on Ello was really interesting: impact of venture capital funding, even for something with well-intensioned founders and chartered as a public benefits corp (PBC). Makes me worried about other PBCs like #posit , though I have no info on their funding or other constraints and so the comparison might be irrelevant
https://waxy.org/2024/01/the-quiet-death-of-ellos-big-dreams/
If you're looking for a new years resolution, might I suggest learning targets in #rstats: https://books.ropensci.org/targets/ . It makes analyses, simulations, and more so much easier. A bit of a learning curve at first, but it is so helpful overall.
@plantarum Agreed it looks weird. The feds say:
"The nursing home data posted reflects cumulative cases and deaths of COVID-19 reported in a facility. A nursing home may report more confirmed COVID-19 cases than beds because they may encounter a large number of admissions and discharges of residents with COVID-19, resulting in more cumulative cases of COVID-19 over time than the number of beds at any one time.
This may also be due to a data entry error on the part of the facility..."
The US government has data on covid incidence and vaccination in nursing homes. It's rather shocking: nearly half of nursing homes have NO healthcare staff fully vaccinated. The data are at https://data.cms.gov/covid-19/covid-19-nursing-home-data, but to help get more attention on this, I launched a quick website visualizing the data:
that allows drilling down to individual facilities.
New species concept just dropped ... no, wait, it actually seems useful. Wayne Maddison & Jeannette Whitton published in @SystBiol's #DiamondOA BSSB so it's free to read and free to authors. Even if the concept itself doesn't take off, I think lots of lectures are going to use the figures and ideas.
https://ssbbulletin.org/index.php/bssb/article/view/9358
#Systematics #SpeciesConcepts #Evolution #Conservation #OpenAccess
Now in #ClimateChange news: the coal-powered electric plant 5 miles from my house has now burned its last chunk of coal as the Tennessee Valley Authority continues moving to clean energy
Was once the largest (by steam volume) power plant in the world.
Interesting package from Norm Matloff on #MachineLearning in #rstats: https://github.com/matloff/qeML.
According to the readme:
"much simpler interface than tidymodels, caret, mlr3, superlearner, SuperML etc."
"includes tutorials on major ML predictive methods, and on special topics such as feature selection and dealing with missing values"
"variety of functions for feature selection and model development"
Client Info
Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst