Laurel Oldach

Laurel Oldach

@laureloldach.bsky.social

Biochemistry, chemical biology & instrumentation at Chemical & Engineering News. I'm just a reporter, standing in front of 2026, asking for some epistemological clarifications. Signal: Laurel_Oldach.07

1.5K followers2.3K following670 posts

Laurel Oldach reposted

Chemjobber
@chemjobber.bsky.social

Do I know any clinical pharmacologists on here? Would you be willing to speak with a high school student? if so, please DM me or email me at chemjobber -at- gmail dot com🧪#chemsky Please RT

Laurel Oldach
@laureloldach.bsky.social

Is there such a thing as a protein? Some proteome researchers, empowered by deeper coverage of more numerous proteomes, are saying no--that proteins don't exist, but diverse proteoforms do. This and other observations from the World HUPO Congress last week: cen.acs.org/biological-c...

cen.acs.org‘Proteins don’t exist’: Scenes from HUPO 2026At the Human Proteome Organization World Congress, researchers turn from building tools to finding out what they can do
Laurel Oldach
@laureloldach.bsky.social

Interesting point about big companies and (relatively) small projects. This makes me wonder what will become of all the drug-repurposing hypotheses the various coscientist systems have been cranking out.

Jamie Cate@jhdcate.bsky.social

Here’s the thing about the Anthropic story. When I worked on biofuels at the EBI funded by BP, it was clear that BP did not have the capacity to think about any product less than 1 MMT (million metric ton for those who use eppendorf tubes). Where does Anthropic see this going?

Laurel Oldach
@laureloldach.bsky.social

Anthropic "CRISPR-like system" opinion: based on where the biochemical/wet-lab validation currently stands as reported in their not-a-preprint, this is a marketing document. The humans involved have made some v. cool mol-bio discoveries in the past, and evidently they expect this to also have ...

View full thread
Laurel Oldach
@laureloldach.bsky.social

Anthropic "CRISPR-like system" opinion: based on where the biochemical/wet-lab validation currently stands as reported in their not-a-preprint, this is a marketing document. The humans involved have made some v. cool mol-bio discoveries in the past, and evidently they expect this to also have ...

View full thread
Laurel Oldach
@laureloldach.bsky.social

But as all of the humans involved know, right now things slow down DRAMATICALLY when you take them from in silico to the wet lab. Which I reckon is why all we've got on this purported novel CRISPR system is a single transcriptomics experiment.

Laurel Oldach
@laureloldach.bsky.social

Anthropic "CRISPR-like system" opinion: based on where the biochemical/wet-lab validation currently stands as reported in their not-a-preprint, this is a marketing document. The humans involved have made some v. cool mol-bio discoveries in the past, and evidently they expect this to also have ...

Laurel Oldach
@laureloldach.bsky.social

... some functional significance. But so far what they've shown is that when you ask Claude for genome mining, it finds novel pattern (which is very cool! I'd have loved to have this as a grad student attempting coevolution analyses!). But as yet, no evidence these sequences do anything besides

Laurel Oldach
@laureloldach.bsky.social

In response to the PR debacle—which, to be fair, probably had more to do with the concurrent announcement by a lead scientist at Anthropic that he was leaving the company because he’d become convinced that the chances AI would kill everyone was unacceptably high—OpenAI launched an advisory group.

Laurel Oldach
@laureloldach.bsky.social

I'm (finally) catching up on the Navier-Stokes proof for work, and wait just one minute here. Is the infinite acceleration of a frictionless-but-fictitious entity really the subject of this AI-resolved proof? really? Please tell me someone who knows math has written a Take on this metaphor.

Laurel Oldach
@laureloldach.bsky.social

Notes on a broadening epistemic crisis:

Marie Walde 🌊🦠🔬@mariescopy.bsky.social

The #microscopy community has serious concerns about this year’s winning #NikonSmallWorld in Motion video. @healthcare.nikon.com, could you please comment on the raw data and acquisition process to clarify whether AI was used? Transparency would help maintain confidence in the competition.

Laurel Oldach reposted

Chemjobber
@chemjobber.bsky.social

From 2022 to spring 2025, American PhD-granting universities posted an average of 289 openings in departments of chemistry and biochemistry. Starting in the fall of 2025, that number fell to 185 openings.

Laurel Oldach
@laureloldach.bsky.social

Key point made here, that prediction markets could- by being publicly-observable indicators of expected success - interfere with clinical trial outcomes. (Also bad for a lot of other reasons.) www.science.org/doi/10.1126/...

science.orgBan bets on unfinished clinical trialsIn July, the prediction market platform Kalshi and data company AppliedXL began a pilot market listing binary, yes-or-no contracts on whether named late-stage clinical trials meet their preregistered ...
Laurel Oldach
@laureloldach.bsky.social

One sure thing they found was that, in their study cohort and period, PFOS + PFOA decrease with age because of a UK restriction/phaseout. Good news, because these contaminants also associate with gene expression & metabolic changes.

Laurel Oldach
@laureloldach.bsky.social

My colleague @bethanyhalford.bsky.social listened to TEN HOURS of medicinal chemistry talks yesterday to bring you the highlights. (Well, I'm told there was lunch, but still.) Spirocycles, deuterated methoxy groups, pyrones: check out these nifty molecules. cen.acs.org/pharmaceutic...

cen.acs.org13 clinical candidates revealed in ChicagoFirst-time disclosures session at ACS Fall 2026 debuted compounds designed to treat cancer, acne, urinary tract infections, and more