Dan MacLean
banner
danmaclean.bsky.social
Dan MacLean
@danmaclean.bsky.social
Head of Bioinformatics at The Sainsbury Laboratory, Hon Professor UEA Computer Science. Likes genomics, stats, proteomics, bioinformatics, R user, Python wrangler. Wiganer.
Pinned
I wrote a handy recipe book for R and Bioinformatics. Available through Packt. www.packtpub.com/en-us/produc...
Reposted by Dan MacLean
LLMs are now widely used in social science as stand-ins for humans—assuming they can produce realistic, human-like text

But... can they? We don’t actually know.

In our new study, we develop a Computational Turing Test.

And our findings are striking:
LLMs may be far less human-like than we think.🧵
Computational Turing Test Reveals Systematic Differences Between Human and AI Language
Large language models (LLMs) are increasingly used in the social sciences to simulate human behavior, based on the assumption that they can generate realistic, human-like text. Yet this assumption rem...
arxiv.org
November 7, 2025 at 11:13 AM
Reposted by Dan MacLean
(With apologies to Randall Munroe)
October 15, 2025 at 7:36 AM
Reposted by Dan MacLean
I really really really liked The Oatmeal’s take on AI here theoatmeal.com/comics/ai_art
A cartoonist's review of AI art - The Oatmeal
This is a comic about AI art.
theoatmeal.com
October 8, 2025 at 1:57 PM
Reposted by Dan MacLean
New pre-print from our group showing remarkable synchrony of proliferative growth and regulated cell death during plant infection- each conidial cell has a different fate - by the rice blast fungus. Led by Alice Eseola with @osesmir.bsky.social Lauren Ryder @danmaclean.bsky.social and Martin Egan
Synchronous spatio-temporal control of autophagy and organelle trafficking is necessary for appressorium-mediated plant infection by Magnaporthe oryzae https://www.biorxiv.org/content/10.1101/2025.10.06.680644v1
October 7, 2025 at 3:29 PM
Reposted by Dan MacLean
Great to hear Burnham's been to Mancwalk 2000.
September 25, 2025 at 9:41 AM
Reposted by Dan MacLean
Normalize never saying "leverage" or "utilize" as a verb. Tempted? "Use" is right there!

Simple words are better words
September 16, 2025 at 5:23 PM
Reposted by Dan MacLean
"It has never been clearer: this is a class war, mediated through one person."
Love this article from @zoesqwilliams.bsky.social. I snorted several times, but she makes some serious points: in the eyes of the tabloids, the working class Rayner can't win, whatever she does.
Why are the tabloids so enraged about every aspect of Angela Rayner’s life? | Zoe Williams
Whether it’s her housing, clothes, the restaurants she goes to, or the fact she watches opera and drinks wine, she can apparently do nothing right, writes Guardian columnist Zoe Williams
www.theguardian.com
September 1, 2025 at 5:01 PM
$ ls -laht
drwxrwsr-x 3 uname grp 27 Nov 16 2006 .nautilus
drwxrws--- 3 uname grp 26 Nov 16 2006 .metacity
-rwxrwx--- 1 uname grp 26 Nov 16 2006 .dmrc

😐
July 31, 2025 at 4:06 PM
Reposted by Dan MacLean
Chris outlining the different AI coding agents that can be used for ontology development. Working on documenting best practices for this ai4curation.GitHub.io/aidocs #BOSC2025 #ISMBECCB2025
July 22, 2025 at 10:59 AM
Reposted by Dan MacLean
Today #BOSC2025 and #BOKR (formerly Bioontologies) COSI’s, Chris Mungall is doing the keynote on Open KBs in the age of generative AI #ISMBECCB2025
July 22, 2025 at 10:32 AM
Reposted by Dan MacLean
*Learning visualization in Python as an R user*
#ggplot
July 22, 2025 at 9:07 AM
Reposted by Dan MacLean
I work with a lot of very intelligent people - the experts in a very select field - yet every morning they fight over the parking spots in the shade. By lunchtime they're all in full sun and my unwanted "sunny" spot keeps my car nice and cool. They never learn.
July 21, 2025 at 7:20 AM
Reposted by Dan MacLean
AlphaFold is great, but contrary to public opinion it has not completely solved the protein folding problem. Much work remains to be done.
clauswilke.substack.com/p/no-alphafo...
No, AlphaFold has not completely solved protein folding
Biology is hard. Yes, even for AI.
clauswilke.substack.com
July 12, 2025 at 6:19 PM
Reposted by Dan MacLean
I recently ran a session at CSAMA with Laurent Gatto and this is the way. "Good enough" with knowledge of the tools that can help you get there. Here's some of the recs from the paper.
July 13, 2025 at 5:13 AM
July 6, 2025 at 9:25 PM
Reposted by Dan MacLean
I told you we’d be back
June 12, 2025 at 4:15 PM
Reposted by Dan MacLean
"AI" - dashare.zone ADMIN
May 14, 2025 at 9:18 PM
Afternoon stroll
May 4, 2025 at 8:42 PM
Reposted by Dan MacLean
"Time is an illusion. Lunchtime doubly so."
April 22, 2025 at 6:45 PM
Reposted by Dan MacLean
#Rstats is being rewritten in Rust, and the R Core already has a book coming out!
April 1, 2025 at 11:03 AM
Reposted by Dan MacLean
The frightening thing about the Vogons was their absolute mindless determination to do whatever mindless thing it was they were determined to do. There was never any point in trying to appeal to their reason because they didn't have any.
March 29, 2025 at 10:40 AM
Reposted by Dan MacLean
"A GO term analysis? how is this helpful?"
March 28, 2025 at 2:18 PM
Reposted by Dan MacLean
This excellent interactive tutorial on misleading data visualizations explores the idea of a "counter chart" — the graph you draw in response to refute a misleading claims

flowingdata.com/projects/dis...
Defense Against Dishonest Charts
This is a guide to protect ourselves and to preserve what is good about turning data into visual things.
flowingdata.com
February 15, 2025 at 6:48 AM
Reposted by Dan MacLean
Plenty of tortured phrases (arxiv.org/abs/2107.0...) in this #IEEE paper. ieeexplore.ieee.org/...

My pubpeer report: pubpeer.com/publicat...
March 20, 2025 at 7:00 AM