Ian Sudbery
iansudbery.bsky.social
Ian Sudbery
@iansudbery.bsky.social
Senior Lecturer in Bioinformatics at the University of Sheffield. Likes gene regulation, 3' UTRs, non-coding RNA and dancing. He/Him/His

Also at IanSudbery@genomic.social
Tip for authors: If you keep telling me how insightful/perceptive/interesting/relevant my comments are in your response to reviewers, I'm not going to think you respect me, I'm going to think you used genAI to write the response.
November 11, 2025 at 11:53 AM
Reposted by Ian Sudbery
yes. being a leftwing ex-Orthodox Jew, I maybe straddle a wider range of views among my friends and family than some (than many?)

it may be shocking to learn that: every bias I have ever seen the BBC accused of, I have also heard someone else say they have the precise opposite one
In response to some of the comments:

If you agree with *all* the reporting of a news organisation that is both independent *and* non-partisan, then it's very unlikely to be independent and non-partisan.

This is true regardless of your politics.

Agreeing with all its reporting is the wrong test
Information underpins democracy and the BBC is a key part of that.

We need to protect our institutions

open.substack.com/pub/christin...
November 11, 2025 at 10:47 AM
Reposted by Ian Sudbery
new Lai lab paper @natsmb.nature.com! Dicer is specifically mutated in cancer, but we don't fully understand its molecular/reg impacts. with @danweihuangfu.bsky.social, we characterized the first knockin Dicer hotspot in hESCs, and found unexpected defects in miRNA biogenesis! 🧬 1/4

rdcu.be/eOc0q
Human DICER1 hotspot mutation induces both loss and gain of miRNA function
Nature Structural & Molecular Biology - Jee et al. study a cancer hotspot allele of DICER1 that disrupts RNaseIIIb activity. Beyond ablating 5p hairpin cleavage, 3p passenger strands are...
rdcu.be
November 10, 2025 at 4:43 PM
Reposted by Ian Sudbery
Picket demonstrating feeling about announcements of cuts to courses at University of Nottingham @uonucu.bsky.social @ucu.org.uk
November 10, 2025 at 9:32 AM
Reposted by Ian Sudbery
This seems likely true from some research we have ongoing as well. One persistent problem the NHS has is it's held to a punishing standard of "efficiency" where it's always running at close to capacity, but any operations researcher will tell you run at 80% or so.

www.bbc.co.uk/news/article...
Will a new mutated flu strain cause a rough winter?
Leading flu experts say they will not be surprised if this year's is the worst flu season for a decade.
www.bbc.co.uk
November 9, 2025 at 7:29 AM
Reposted by Ian Sudbery
Are you using any of our factor models, such as MOFA? 🛵
You might’ve found it challenging to tailor them to your specific use cases - not anymore!

Introducing MOFA-FLEX: a flexible, modular factor analysis framework designed for customizable modeling across diverse multi-omics data scenarios. 1/n
November 7, 2025 at 10:29 AM
Reposted by Ian Sudbery
Staff at Dundee University are taking action again next week to defend jobs and resist compulsory redundancies.

A year after the university’s financial crisis, uncertainty still hangs over hundreds of staff.

Solidarity with Dundee UCU members fighting for fairness and job security. ✊
News that further job cuts are coming is just another shocking revelation in a year when staff have been told one thing only to be told something different over the number of jobs to be cut and the need for compulsory redundancies.

DUCU co-president, Melissa D’Ascenzio

news.stv.tv/north/dundee...
University staff strikes set to continue amid job cut threats
Further strike action is expected from UCU members from November 10 next week amid a long running dispute with university bosses.
news.stv.tv
November 5, 2025 at 11:10 AM
Reposted by Ian Sudbery
Chris' lab does very cool work, can recommend highly.
November 5, 2025 at 9:27 AM
Important to note that this is 3'-end sequencing, not full transcript sequening, so no use to anyone interested in transcript-structure, splicing, unannotated genes, or non-polyA transcripts.

But still, whole genome gene-expression, from cell pellet samples in 3 days for $50 is pretty amazing.
🦖 Something HUGE just hatched.

Plasmidsaurus now offers RNA sequencing for gene expression analysis:
• As fast as 3 day turnaround
• $50/sample for academia, $80 for industry
• Up to ~10M unique transcript 3’ end reads per sample
• Interactive results

Explore Plasmidsaurus RNA-Seq today.
November 3, 2025 at 1:32 PM
Reposted by Ian Sudbery
Next week LinkedIn will use your personal data for AI training by automatically enabling permissions.

To manually turn off go to:

Settings ➡️ Data Privacy ➡️ under
“How LinkedIn uses your data” click “Data for Generative AI improvement” ➡️ toggle off
November 2, 2025 at 5:07 AM
Reposted by Ian Sudbery
this is so fucking bad.

men get breast cancer. cis men get breast cancer. cis men who don't have Kleinfelter's can also get breast cancer.

bodies are complicated, that's the thing. stuff is *complicated*.
As a male breast cancer survivor, I have a two word response to this and the second word is "you" and the first word is a word I try not to say too often publicly because I write and draw kids comics.

But yeah... those two words. In bold print. And large letters.
SCOOP

The admin is making it harder for male vets w/ breast cancer to get their care covered

The admin cites Trump order: “Defending Women from Gender Ideology Extremism and Restoring Biological Truth to the Federal Government”

I got the memo laying it all out

www.propublica.org/article/vete...
October 29, 2025 at 12:27 PM
Reposted by Ian Sudbery
After 13 years of analyzing sequencing data, I am an "expert" in doing it.
I gained those experiences not because I am smarter,

but simply because

I made more mistakes,

and encountered more problems.
October 29, 2025 at 1:45 PM
Reposted by Ian Sudbery
Often forgotten, ignored, or overlooked detail of peer review practices:
While there may be "pay to publish" and "predatory" journals, the fact that an editor does not agree with the position of any given peer reviewer is not evidence of this. Peer review is merely advisory, it is not decision making. Confusion over this leads to some very bad reviewer practices.
As far as I see it, it is pay to publish. They have ignored my review comments before just went forward publishing a paper that was simply trash quality. The quality of some of the papers are very bad, which makes me question the editorial and review process. There are of course many good papers.
October 27, 2025 at 4:35 PM
Graeber outlined 5 types of BS job: Flunkies, Goons, Box-tickers, duct-tapers and Taskmasters.

Ask yourself: Which of these could be automated?
October 27, 2025 at 1:44 PM
Reposted by Ian Sudbery
thread of skulls, which apart from being a useful guide to Nazi/not Nazi skulls has an interesting cultural angle.
Apparently some folks are playing real dumb about skull symbols so let’s play a little game I like to call “Nazi skull or not a Nazi skull”

First up, the Jolly Roger. This is a pirate skull, not a Nazi skull. It means you might get robbed, but prolly not genocided.

With me so far? 1/x
October 26, 2025 at 7:15 AM
Reposted by Ian Sudbery
New blog post alert! A personal retrospective on reaching a decade of development for @multiqc.info 😊
🎉 Celebrating 10 years of @multiqc.info! 📊 The numbers speak volumes:

🔹 Support for ~170 bioinformatics tools
🔹 1 run every 3 seconds (25k+ daily runs)
🔹 From single contributor to 235 contributors
🔹 7000+ journal citations (~8-9 per day)

📚 Discover a timeline of innovation:
hubs.ly/Q03PsbTP0
October 24, 2025 at 4:17 PM
Reposted by Ian Sudbery
I think I just found my first AI generated preprint on bioRxiv. Unclear methodology, invalid citations, and dead links. Are these supposed to get reported somewhere?
October 20, 2025 at 2:36 PM
Reposted by Ian Sudbery
This remains the funniest way to hear about an internet outage, though.
October 20, 2025 at 8:41 AM
Reposted by Ian Sudbery
I regularly get derailed when someone tries to benchmark clustering methods, trying the find the "best" one or the one that agrees with "ground truth".

Which clustering of the below animals is the best or the true one? By row, or by column?
October 16, 2025 at 9:44 AM
Reposted by Ian Sudbery
Hi bioinformatics, genomics and CS friends! Please help me spread the word. I'm hiring a postdoc! Come work on cutting edge method development in algorithmic genomics with me and my group at @umdscience.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!
October 10, 2025 at 1:02 PM
Reposted by Ian Sudbery
15 years in the making, we confirmed that mitochondria - the powerhouse of the cell - have an unusual localization in patients who experience psychosis (including schizophrenia and bipolar disorders). You’ll never guess what kind of patient cells we used to make this discovery… 🧵
October 10, 2025 at 4:47 PM
Reposted by Ian Sudbery
RIP John Gurdon
October 7, 2025 at 6:06 PM
Reposted by Ian Sudbery
RIP. He changed Dev Bio in so many ways.
October 7, 2025 at 4:43 PM
Reposted by Ian Sudbery
This year's Crick @crick.ac.uk PhD programme is open for applications

- £27,715 tax-free student stipend per year for 4 years
- Tuition fees paid
- Student visa fee & health surcharges covered

Deadline 5 Nov

We have a project available. For all the details visit:

www.crick.ac.uk/careers-stud...
Briscoe Lab | Cellular Dynamics and Pattern Formation in Vertebrate Development
www.crick.ac.uk
October 6, 2025 at 11:37 AM
Reposted by Ian Sudbery
These are hilarious.
October 5, 2025 at 8:51 AM