Lightnews — Scholar-powered news

John P. Overington

@jpoverington.bsky.social

900 followers 1.4K following 100 posts

Work at drughunter.com - Drug Discovery, Data Science, Cheminformatics, Computational Medicinal Chemistry, Drug Repositioning, Structural Bioinformatics, CADD, SaaS, DaaS, AI/ML https://scholar.google.com/citations?user=52VMFeYAAAAJ

scholar.google.com

Posts Media Videos Starter Packs

John P. Overington @jpoverington.bsky.social · 6d

I was 13. The swearing on Plaistow Patricia, the wonderful double entendres and the jolly pathos. My Old Man, I still cry at. Cuts deep.

John P. Overington @jpoverington.bsky.social · 8d

Apparently it’s 48 years ago since New Boots and Panties was released. A true classic album, survived the test of time for sure!

1 3

John P. Overington @jpoverington.bsky.social · 12d

The now legend of the gluing of the album covers, and the surrealist concept of destroying other records in the store using the cover, I think it is unmatched

John P. Overington @jpoverington.bsky.social · 12d

For self conscious pseudo-intellectual teenagers from post-industrial Britain in the 1980s this is big news. For the rest of the world - nothing to see here.

1 2

Reposted by John P. Overington

Andrew White 🐦‍⬛ @andrew.diffuse.one · 13d

I finished my estimate on required compute to make an atomic-resolution virtual cell: 10^38 FLOPs to simulate a human cell for 1 day. We should be able to do this simulation in 2074 using 200 TW of power. 1/3

4 3 11

John P. Overington @jpoverington.bsky.social · 14d

What’s wrong with wc on the command line?

John P. Overington @jpoverington.bsky.social · 15d

1 1

John P. Overington @jpoverington.bsky.social · Sep 7

Nostalgia ain’t what it used to be. Just found a microfiche. Big stack of these. If only I had the reader. I bet they’re museum pieces now.

3 6

John P. Overington @jpoverington.bsky.social · Sep 7

Just sorting through some old files.
Really fun to step back in time. This must have been 1988 or so. Acetates, permanent markers. None of your AI assistants in putting this together. And it shows 😂😂

John P. Overington @jpoverington.bsky.social · Sep 1

As a Brit expat in the States, it’s sad to see that access to the BBC News website is now paywalled. $90 a year!

2 1

John P. Overington @jpoverington.bsky.social · Aug 22

It’s August in our home, and that means getting ready for Halloween!! The theme this year (apparently) is Labyrinth - here is David Bowie.

John P. Overington @jpoverington.bsky.social · Aug 14

A sort of nerdy question around licenses. Imagine I subscribe to a journal in which they don’t allow text mining on the content. Seems fair. But if is store the pdf locally, then my Mac will index it and allow full text searching. Does this count as text mining?

John P. Overington @jpoverington.bsky.social · Aug 6

Congratulations!! I heard that a metric to aim for is an H-index that matches your age; hang on, or was it your waist size?

1 1

John P. Overington @jpoverington.bsky.social · Aug 5

Getting frustrated with the number of resources that disappear, change terms of use, etc. Seems cheeky that people publish in places like NAR Database with an 'open' resource, then flip to closed. Tempted to put together a table of all the NAR DB resources and plot existence/current license....

1 4

John P. Overington @jpoverington.bsky.social · Jul 20

And while I’m thinking of Nature, their typography has gone downhill. The spacing between words in titles is now atrocious! Defense against OCR LLM tokenization or just poor visual design? 🤣

1 3

John P. Overington @jpoverington.bsky.social · Jul 20

More seriously though, have there ever been any scholarly reviews of reliability of patent literature for fakery. I realise there are loads of batshit crazy teleportation patents, so restricting to credible applicants/areas should be easy to do.

John P. Overington @jpoverington.bsky.social · Jul 20

Bit of an insult for Nature magazine in the Google response to asking if fake patents and fake publications are different.

Reposted by John P. Overington

Stephen Curry @scurry.bsky.social · Jul 16

The times they are a changin’.

RCSB Protein Data Bank @rcsbpdb.bsky.social · Jul 16

Transitioning to PDBx/mmCIF and Extended PDB IDs
PDB users are strongly encouraged to make changes to software; use the full 12-character ID in all communications; and encourage others to adopt ASAP

Details at wwPDB: www.wwpdb.org/news/n...

1 2 5

John P. Overington @jpoverington.bsky.social · Jul 14

Shucks! I’m blushing!!

Reposted by John P. Overington

Roland Dunbrack 🏳️‍🌈 @rolanddunbrack.bsky.social · Jul 2

CASP is the main reason the protein structure prediction technology and research field advanced over the last 30 years. And the main reason AI based methods have been accepted and widely applied in biology. So shortsighted of NIH to postpone or even halt funding. John Moult is a scientific hero.

Science Magazine @science.org · Jul 2

Exclusive: An international scientific competition widely credited with spurring the development of artificial intelligence for biology appears to be on its deathbed. scim.ag/44ukS90

Exclusive: Famed protein structure competition nears end as NIH grant money runs out

Agency silent on funding renewal for contest that inspired creation of AIs that predicted how proteins would fold

scim.ag

1 14 41

John P. Overington @jpoverington.bsky.social · Jul 2

There is this table of spelling differences from Clarivate webofscience.help.clarivate.com/en-us/Conten... not sure why pantyhose and pajamas made the science list though.

webofscience.help.clarivate.com

John P. Overington @jpoverington.bsky.social · Jul 1

That's it. It's done. I am no longer on the other place. I'm going to miss some of the science on there, and in particular the biotech analyst community seems stuck there, but enough was enough. It had to go.

John P. Overington @jpoverington.bsky.social · Jul 1

Thanks, it works to some extent on specific examples, but I want a fairly comprehensive list to help with automated canonicalization/localisation of a large number of text documents. Also in search interface implementation.

John P. Overington @jpoverington.bsky.social · Jul 1

Is there anywhere online a free lexicon for alternate spelling of scientific terms between US and UK English. Something that contains the equivalence of sulphotransferase = sulfotransferase, protease = proteinase, etc. But also smart to know that synthase and synthetase are different concepts.

John P. Overington @jpoverington.bsky.social · Jul 1

UI oddity of the day - a box to ask if I'm human, which looks like the only option I have is to click to say no, turns out it's a slider hiding as a tick box.