Travis Wheeler
LightNews LightNews
banner
wheelerlab.org
Travis Wheeler
@wheelerlab.org
340 followers 200 following 270 posts
Associate Professor, University of Arizona
Algorithms, Machine Learning, Data Resources
Computational Genomics, Drug Discovery, Animal Tracking, etc.
wheelerlab.org
Posts Replies Media Videos
wheelerlab.org
Math is hard
September 29, 2025 at 2:46 AM
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
September 12, 2025 at 8:36 PM
wheelerlab.org
No more funny-looking than the top dog in my kitchen
Picture of a Vizsla in a costume, sitting on another dog (who is not in a costume, and seems unperturbed by the whole "I'm being sat on" thing)
August 30, 2025 at 11:00 PM
wheelerlab.org
I confirm your opinion
August 2, 2025 at 10:12 PM
wheelerlab.org
A Representation Learning approach to virtual ligand-screening. Produces good enrichment for active ligands, and can scan billions of candidate small molecules in a few hours..

Simpatico: www.biorxiv.org/content/10.1...
June 20, 2025 at 2:47 PM
wheelerlab.org
I am definitely NOT burning the candle at both ends
Generated image of a very complicated multi-head candle, with all ends burning in a mighty conflagration. In the background (blurred), is a bewildered-looking man watching the candles burn.
June 9, 2025 at 9:08 PM
wheelerlab.org
Interesting read -- I'd have agreed whole-heartedly when it was written. Starting to feel a little dated:
Received on 05 February 2025;
June 4, 2025 at 6:43 AM
wheelerlab.org
I guess she just really wants to be a GOAT (Matthew 25:31-46):
A quote from Matthew 25:41–46: (The people on the left are the goats) 41 “Then he will say to those on his left, ‘Depart from me, you who are cursed, into the eternal fire prepared for the devil and his angels. 42 For I was hungry and you gave me nothing to eat, I was thirsty and you gave me nothing to drink, 43 I was a stranger and you did not invite me in, I needed clothes and you did not clothe me, I was sick and in prison and you did not look after me.’ 44 “They also will answer, ‘Lord, when did we see you hungry or thirsty or a stranger or needing clothes or sick or in prison, and did not help you?’ 45 “He will reply, ‘Truly I tell you, whatever you did not do for one of the least of these, you did not do for me.’ 46 “Then they will go away to eternal punishment, but the righteous to eternal life.”
June 1, 2025 at 12:08 AM
wheelerlab.org
I'm old enough to remember when Reagan was considered a conservative. This detestable description of our nation (below) proves that he was actually a Marxist, hellbent on destroying our great nation.

www.reaganlibrary.gov/archives/spe...
A quote from Reagan's farewell address: "I've spoken of the shining city all my political life, but I don't know if I ever quite communicated what I saw when I said it. But in my mind it was a tall, proud city built on rocks stronger than oceans, wind-swept, God-blessed, and teeming with people of all kinds living in harmony and peace; a city with free ports that hummed with commerce and creativity. And if there had to be city walls, the walls had doors and the doors were open to anyone with the will and the heart to get here. That's how I saw it, and see it still."
May 29, 2025 at 5:37 PM
wheelerlab.org
OMG, @deltahelp.bsky.social ... I'm trying to change a flight for my kid's trip to Europe. It's a change from a flight that was ~$600 to one that currently lists at $670. So maybe it makes sense to charge me $70. Maybe it makes sense to charge me some extra? But, uh ... $4,000 ???!!! JFC
This image captures a confusing airline pricing scenario for a flight itinerary, with annotations highlighting the issue. Here’s what’s happening: 1. Flight Itinerary: • The itinerary shows a one-way flight from Los Angeles (LAX) to Cardiff, Wales (CWL) via Minneapolis (MSP) and Amsterdam (AMS), departing on Sunday, July 20 at 9:00 am and arriving at 9:15 am the next day. • The flights are operated by Delta (with codeshare partner German Airways for the final leg). 2. Price Display: • On the left side of the image, it shows a Delta Main Economy Light fare of $670 for the one-way trip. • Next to it, Delta Comfort Europe Economy is shown at $941. 3. Price Discrepancy: • On the right-hand side, a message box from the airline’s customer support or booking system is shown. • It states that the “new total cost per passenger, including all taxes and carrier-imposed fees for this itinerary, is $5,151.31, which is an additional $3,986.60 from the original fare of $1,164.71.” 4. Annotation: • There are orange circles and question marks drawn around the $670 fare and the dramatic difference highlighted in the support message.
May 28, 2025 at 4:40 AM
wheelerlab.org
I don't think it's that. I paste the full url (or even just: globalfishingwatch.org), and it (usually, but not always) presents me with that "request is being verified" thing, then a 404. I think that green text means Cloudflare is over-protecting your site (see image). Maybe I'm just unlucky.
May 18, 2025 at 3:48 AM
wheelerlab.org
Strange. I click the link, and I get sent to a page with "please wait while your request is being verified" (see image) then a really gross URL (see 2nd image)
May 17, 2025 at 10:22 PM
wheelerlab.org
On second thought, we should be "Cautious, concerned, but **not dismissive**."

So now we know how we should feel ...
April 2, 2025 at 7:45 PM
wheelerlab.org
ChatGPT thinks there are potential concerns ... so we're all on the same page! (but wait, there's more)
You’re pointing at a really important issue here, and honestly, it’s something that’s starting to worry a lot of researchers, educators, and readers. The image you shared is a perfect example of AI-generated summaries of scientific studies that are being presented without formal review or input from the original authors. Here’s why this can be problematic: Potential Concerns 1. Accuracy & Nuance Loss AI-generated summaries often simplify complex findings. In fields like ecology, genomics, or clinical research, subtle conclusions or limitations may get lost, misinterpreted, or overstated. 2. No Accountability or Peer Validation Unlike the peer-review process, these AI outputs don’t undergo human validation. Readers may assume the summaries carry the same scientific rigor as the original paper, which is misleading. 3. Bias & Misrepresentation AI models can subtly (or not so subtly) introduce biases based on the data they were trained on. They may emphasize certain results over others, or misinterpret statistical findings. 4. Undermines Authoritative Voice It circumvents the voice and intent of the authors and reviewers, who are the experts responsible for contextualizing the data. 5. Erosion of Scientific Literacy Readers who rely only on these “quick answers” without reading the paper may develop an incomplete or distorted understanding of the research.
April 2, 2025 at 7:42 PM
wheelerlab.org
March 24, 2025 at 1:19 AM
wheelerlab.org
I think I'll just keep on meeting with you. I think understand things that gpt doesn't.
screen shot of chatgpt session. My prompt ------------ interpret this comic, then come up with a witty reply The reply ---------- Interpretation: The comic pokes fun at our tendency to offload tasks—especially mundane ones like meetings—to AI, only to ignore the results anyway. It’s highlighting that sometimes all the planning and summarizing ends up feeling pointless, because we never actually use the information we requested. Witty Reply: “Brilliant plan! Let’s schedule an AI-led debrief so we can promptly forget to read that summary too.”
March 1, 2025 at 8:27 PM
wheelerlab.org
Cartoon image of a female saying "Aye Aye Captain" Another image of a female saying "Aye Aye, Captain".
February 10, 2025 at 9:05 PM
wheelerlab.org
This you?
Anshul Kundaje @anshulkundaje.bsky.social · 2mo Pinning this for myself: I am determined to be kinder, gentler on this site compared to the netherworld we left – more like how we behave in the real world. Will still have a lot of (ideally constructively) critical things to say & have plenty of arguments especially on controversial topics.
February 4, 2025 at 5:31 PM
wheelerlab.org
So now I need to document my work with tattoos?
A photo of Leonard, the main character form the movie "Memento". It shows him shirtless, with a bunch of tattoos on his chest and arms.
January 30, 2025 at 6:10 PM
wheelerlab.org
Ah, the good old days ...
A screenshot showing that pubmed currently (as of 3:30PM EST) returns a 500 internal service error.
January 28, 2025 at 8:36 PM
wheelerlab.org
The math checks out. The logic is, um, checkered.
(Question) Is 12 teaspoons equal to 1/4 cup? (Answer) No, 12 teaspoons is not a 1/4 cup. Here’s the breakdown: • 1 tablespoon = 3 teaspoons • 1/4 cup = 4 tablespoons Thus: So 12 teaspoons is equal to 1/4 cup.
January 12, 2025 at 8:29 PM
wheelerlab.org
There once was a writer who said,
“AI is dull as cold bread.”
But the bots, full of sass,
Said, “You’d better not pass,
Till you’ve laughed at the rhymes in your head!”

(This make me nervous)
January 9, 2025 at 11:55 PM
wheelerlab.org
We're still hunting for a real logo, but for now I'm going with this: it's a picture of a nail, with a ribbon diagram of P93604 draped over it. You can probably guess what language we used to implement nail.
github.com/TravisWheele...
A picture of a rusty nail, with an image of a protein structure draped over the nail. The protein is P93604 (Rust resistance kinase). That's incredibly funny because nail is implemented in the programming language called Rust. My children don't think it's funny, but they're wrong.
December 25, 2024 at 6:30 PM
wheelerlab.org
A Christmas present for those who like fast and sensitive protein annotation tools!

This shows runtime vs recall on a test with ~1400 query MSAs, with ~20 true matches per query planted among 2M decoy sequences. Speed gains over H3 grow for bigger datasets, and scale linearly with MMSeqs2 runtime.
A plot showing a tradeoff between recall and runtime. It shows that nail is roughly 20x faster than HMMER ~2-3x slower than MMSeqs2), and gains ~15% of the ~25% gap in recall between MMseqs2 and HMMER3
December 25, 2024 at 6:24 PM
wheelerlab.org
I made that shift early in my first winter in Montana.
Travis in a very warm ski mask.
December 25, 2024 at 6:09 PM