Travis Wheeler

I talk a lot about Rust for building high-perf (& even non-perf critical) software, & scientific software in particular. I often discuss what's interesting to me, but wanted to offer the chance to those interested for me to answer their questions about Rust in science. Fire away with questions!🧬🖥️

Know an Evolutionary Genomicist looking for a faculty position? Join us in EEB @utknoxville.bsky.social Position is open for studying any organism, but personally I have some botany bias. 🌱 Apply before Sept 19 for full consideration. apply.interfolio.com/170735 Please share widely. Thanks!

I confirm your opinion

My first first author manuscript has posted! I’m very excited to share SWIF-TE with yall. It is a fast, memory efficient tool to identify novel TE insertions from short-read data. 🧬 #TEworldwide #transposons

we asked a simple question:
What does it take to learn the rules of RNA base pairing?

using standard deep-learning technics, got a simple answer:
don't need structures, nor alignments or many parameters
only a few RNA sequences and 21 parameters;
doi.org/10.1101/2025...

Watching @jonhoo.eu's vibe-coding stream exactly matches my personal experience using agentic AI for programming. That is, it's fine for things that are basiclly pattern matching, but fails quite badly (i.e. is slower than just doing it manually) for changes that require reasonable thought.

\toprule & & \multicolumn{4}{c}{CPU time / h} & & \multicolumn{4}{c}{Max RSS\tnote{a}{ } / GB} \\ \cmidrule(lr){3-6} \cmidrule(lr){8-11} Read length / bp & & 24 & 100 & 200 & 24--200 & & 24 & 100 & 200 & 24--200\\ \midrule Newmap & & 4 & 9 & 16 & 150.5 & & 7.6 & 7.6 & 7.6 & 7.6 \\ Genmap & & 2 & 1 & 1 & 275 & & 13.1 & 13.1 & 13.1 & 13.1 \\ Umap & & 31 & 42 & 57 & >4752\tnote{b} & & 7.8 & 7.5 & 5.9 & ${\sim}7.8$\tnote{c}\\ \bottomrule \end{tabular} \begin{tablenotes} \item[a] Maximum resident set size. \item[b] Estimate from a linear extrapolation of \qty{24}{\basepair} CPU time. \item[c] Estimate from the largest maximum RSS found across all Umap benchmarks. \end{tablenotes}

The new version is built on @wheelerlab.org's AvxWindowedFMindex and some other tricks we came up with, and is *so fast*.

Umap and Bismap: quantifying genome and methylome mappability Abstract. Short-read sequencing enables assessment of genetic and biochemical traits of individual genomic regions, such as the location of genetic variati

We are creating new software to quantify genome mappability, a successor to Umap, our previous work described in the linked paper. What standard read lengths should we provide in BED/Wiggle format? Previous was 24/36/50/100 bp. academic.oup.com/nar/article/...

GitHub - lh3/longdust: Identify long STRs, VNTRs, satellite DNA and other low-complexity regions in a genome Identify long STRs, VNTRs, satellite DNA and other low-complexity regions in a genome - lh3/longdust

lh3/longdust: Identify long STRs, VNTRs, satellite DNA and other low-complexity regions in a genome github.com/lh3/longdust 🧬🖥️🧪

FAMSA2 enables accurate multiple sequence alignment at protein-universe scale We introduce FAMSA2, an algorithm that produces high-accuracy multiple protein sequence alignments with unprecedented speed. Across structural, phylogenetic, and functional benchmarks, FAMSA2 matches ...

Interested in a tool that aligns millions of proteins in minutes with quality similar to or better than the state-of-the-art utilities? Please take a look at our FAMSA2 paper: www.biorxiv.org/content/10.1...
and GH repo: github.com/refresh-bio/...

Really important points by @alexbateman1.bsky.social on the importance of curation, addressing some of the myths out there.

Deeply resonates with the commentary Paul Thomas and I wrote last year www.nature.com/articles/s41...

it's longer than the drive from ANY house in Missoula to the Missoula airport.

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...

Exclusive: Waymo rides cost more than Uber, Lyft — and people are paying anyway | TechCrunch A new analysis done by ride-hailing aggregator Obi shows Waymos cost more especially on shorter trips. They also have longer wait times.

Oh, I get it: their rides cost ... way mo ... than the other guys.

That is 54, 5-year NIH biomedical research grants.

Or 270 years worth of biomedical research.

YouTube video by wjteegarden Now listen to this carefully, Norman: I am lying...

www.youtube.com/watch?v=4WRt...

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.

I am definitely NOT burning the candle at both ends

Hello bluesky world! Newbee here! I have a postdoc position immediately available in my lab. It will focus on identifying high-quality transposons in many genomes and finding their impacts in evolution and traits. Most works, including EDTA2 development and annotation of 400+ genomes, are done! 1/n

@googleworkspace.bsky.social :

Is this ***yet another*** in an endless line of features that non-paying customers get to use, but I can't access because I have a paid workspace account? I'm running out of room to keep track of them all:

blog.google/products/map...

👀 👀 👀

Some of us look for ways to develop a joke over the course of a paragraph. Others just look for the quick win with a responsive quip.

But really: everybody's looking for something.

Very excited to share I’ll be starting my own group at University of Utah in the Department of Human Genetics in the new year! Reach out if you are interested! vollgerlab.com

Received on 05 February 2025;

Interesting read -- I'd have agreed whole-heartedly when it was written. Starting to feel a little dated:

CATH-Gene3D annotation visualised on BFVD structure prediction of UniProt A0A075FDU7

🦠🧬Exciting news! BFVD predicted viral protein structures are now available in @InterProDB!
For the first time, explore InterPro functional annotations directly on 3D structures of viral proteins not covered by AlphaFold DB. This bridges a critical gap in structural virology research.

Howard's: evolution of microbial genes, genomes and communities. Can be experimental or computational. If interested, contact Howard Ochman, howard.ochman@austin.utexas.edu
Homepage: web.biosci.utexas.edu/ochman/index...

An update on open postdoc positions, in my group and the adjacent group of Howard Ochman, here at UT-Austin. #SymbioSky

Mine: insect-bacterial symbiosis with emphasis on intracellular endosymbionts and molecular mechanisms.
web.biosci.utexas.edu/moran/contac...

mdrepo.org is still live and kicking, and still accepting new contributions.

Phage people: Rich Losick and I are combing the world looking for T4 rIIB mutant FC0 (also known as P13). FC0 was the starting point for Francis Crick's beautiful 1961 paper on the triplet nature of the genetic code. We want to sequence it. Anyone have it in an ancient stock collection?

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.”

I guess she just really wants to be a GOAT (Matthew 25:31-46):

I was 100% certain that this was a deepfake, 'cause ... I mean, do I need to explain? But, nope, it's the real deal:

thehill.com/homenews/sen...