Travis Wheeler
@wheelerlab.org
330 followers 200 following 260 posts
Associate Professor, University of Arizona Algorithms, Machine Learning, Data Resources Computational Genomics, Drug Discovery, Animal Tracking, etc. wheelerlab.org
Posts Media Videos Starter Packs
wheelerlab.org
Dream postdoc - Rob's science is excellent (and he's pretty great, too)
robp.bsky.social
And it's posted! If you're interested and eligible, please consider applying through the UMD portal: umd.wd1.myworkdayjobs.com/en-US/UMCP/j....

If you're a PI working in algorithmic genomics (& you can recommend my lab to your top graduating students ;P), please let them know!
wheelerlab.org
Important work being done here.
vvvvvvvvv
vvvvv
vv
paulisci.bsky.social
A Brief History of Men are Becoming Less Manly

🧵
wheelerlab.org
Quite the complement.
wheelerlab.org
I think that's "juice of a lemon".
Reposted by Travis Wheeler
wheelerlab.org
TIL: 10^60 is far greater than 10^67
wheelerlab.org
So you're saying there's a chance that 10^60 is "far greater than the number of atoms in the Milky Way"?

This restores my faith in Google AI answers - thanks!
wheelerlab.org
TIL: 10^60 is far greater than 10^67
wheelerlab.org
Look on the bright side. With AI's help, there's actually a chance you could run a 2:10:00 marathon, since you might actually be able to hold an 5:00/km (8:03/mile) pace. That's incredible news!
greglandrum.bsky.social
It's great to see that the piles of money and enormous amounts of electricity and other resources used to train these models has produced such stunning and useful results. And I'm SUPER glad that google now always shows me this garbage.
A screenshot showing the AI Overview result for the google query. "What is the required pace for a 2:10 marathon"
The AI-generated answer is egregiously wrong.
Reposted by Travis Wheeler
otagobiochemist.bsky.social
We're looking for a new lecturer, here at the Department of Biochemistry in Dunedin, New Zealand. "Applications are invited in all areas of biochemistry, molecular genetics, and bioinformatics." otago.taleo.net/careersectio...
Reposted by Travis Wheeler
wheelerlab.org
I share this picture of a printed out genome every once in while.

It just occurred to me: it's got to be cheaper to acquire a finished copy of your genome (high accuracy, phased, SV-resolved) than to print it as a collection of bound books.

(back-of-envelope checks out - anyone done the math?)
A photo of the first printout of the human reference genome presented as a series of books, displayed at the Wellcome Collection, London.

Image taken from https://en.wikipedia.org/wiki/Reference_genome
wheelerlab.org
Everyone knows you can skip the middle part of a book and get most of the story.
wheelerlab.org
I share this picture of a printed out genome every once in while.

It just occurred to me: it's got to be cheaper to acquire a finished copy of your genome (high accuracy, phased, SV-resolved) than to print it as a collection of bound books.

(back-of-envelope checks out - anyone done the math?)
A photo of the first printout of the human reference genome presented as a series of books, displayed at the Wellcome Collection, London.

Image taken from https://en.wikipedia.org/wiki/Reference_genome
wheelerlab.org
It's strange to think how many times I've asked myself this very question
surtlab.bsky.social
Wait what????

If you’ve ever wondered how you could learn about cool science but also answer the question “what was the town like where Dave grew up”…then this meeting is exactly for you
danielbolnick.bsky.social
The next stand-alone meeting of the American Society of Naturalists @asn-amnat.bsky.social will be on the East Coast!!! Mark your calendars for January 8-12, 2027 and we'll talk evolution, ecology, behavior, and integrative organismal biology at www.themansionatglencove.com/meeting-venues
wheelerlab.org
After some lengthy embargo period (like: 5 or 10 years). Cool.

Next year. Abso-f-ing-lutely terrible.

The scenario: I have an idea. Reviewer #3 torpedoes it. I need to wait months to submit again. Someone else scoops up the ideas and writes a similar proposal while I'm in the next cycle.
wheelerlab.org
slightly diminishing a book:

Kitty's Cradle
wheelerlab.org
slightly diminishing a book:

The Jogging Man
andreapitzer.bsky.social
slightly diminishing a book:

Mademoiselle Bovary
wheelerlab.org
slightly diminishing a book:

1983
wheelerlab.org
slightly diminishing a book:

The Jogging Man
andreapitzer.bsky.social
slightly diminishing a book:

Mademoiselle Bovary
wheelerlab.org
slightly diminishing a book:

The Jogging Man
andreapitzer.bsky.social
slightly diminishing a book:

Mademoiselle Bovary
booktweeting.bsky.social
slightly diminishing a book:

The Tale of Two Towns
wheelerlab.org
As well-stated as anything I've seen about the problem with "using AI to help me write"
waynemaddison.bsky.social
Confused writing is usually a symptom of confused thinking. As we struggle to clarify writing, we clarify our thoughts. AI writing aids rob us of that struggle, leaving clean-looking text and thoughts still confused for lack of inspection. Writing is not just a product; it is a diagnostic tool.
wheelerlab.org
MetaboLights, I think
wheelerlab.org
Project in collaboration w/ @networksunknown.bsky.social.

Goal: improve identity continuity in multi-animal tracking.

Mix pose prediction from AI computer vision methods with classical Markov processes, add a dash of path pruning and a new UI for efficient error correction: diplomattrack.org
wheelerlab.org
They were also bred to willingly settle into dog piles. The breeding is strong enough that they're unhappy if they're not sitting on another dog.

They were also bred to wear costumes when appropriate, I think. (I'll need to look that one up)