iiinova — where your digital self keeps evolving

Reposted by Jeff Lewis

Mark Peifer (He, him) @peiferlabunc.bsky.social · 2h

One of the coolest topics in molecular evolution

Reposted by Jeff Lewis

Pedro Beltrao @pedrobeltrao.bsky.social · 2h

Great point made on the other site. The OpenAI math results can't even be cited. Deepmind has kept a tradition of publishing their results but OpenAI and Anthropic have been reporting these as repos and press releases.

151

Jeff Lewis @lewislab.bsky.social · 2h

At least for latest (controversial) bio release, there is an associated preprint. Strongly agree that publishing needs to be the norm.

100

Reposted by Jeff Lewis

Mark Histed @markhisted.org · 10h

Umm wow. Michael Lewis’s new book says what about NIH??? “He was terrified," said Brought. "He'd never been close to power. He'd never even worked with someone who had been so close to power… He was there because he’d follow orders.”

the job, Memoli simply did whatever downtown told him to do. And downtown had told him to do crazy stuff: fire all the younger scientists, for example, and forbid NIH scientists from buying any supplies that cost more than a dollar.
Memoli's executive secretary, a former Marine Corps intelligence officer named Nate Brought, had finally sat Memoli down and said, Right now you kind of look like a bitch. And everyone here is seeing you like a bitch. You gotta man up. "He was terrified," said Brought. "He'd never been close to power. He'd never even worked with someone who had been so close to power. He didn't know what he was doing. He was there because he'd follow orders."

1377

Jeff Lewis @lewislab.bsky.social · 2h

Spinelessness seems like the charitable interpretation. Alt hypothesis is that he’s a true believer.

000

Reposted by Jeff Lewis

Maitreya Dunham @maitreya.bsky.social · 16h

While book chapters are frequently a bear to write, it’s always thrilling to get a physical object at the end. Thanks to my coauthors Chris and Barbara! We got the intro to yeast chapters for this nice volume from MBAA.

3242

Reposted by Jeff Lewis

Alex Merz 🇺🇸🇨🇦🇺🇦 @merz.bsky.social · 06/10/2026

I just finished reading the academic paper that this news article briefly describes. It's terrific, and anyone teaching molecular biology might benefit from reading it. doi.org/10.1007/s107...

44612

Reposted by Jeff Lewis

I-75 Scientist @flscitriguy.bsky.social · 06/10/2026

NIGMS is soliciting input from the Institutional Development Award (IDeA) program community to help determine the best timelines for the submission and review of IDeA program post-award project (such as pilot and research project) prior approval requests. grants.nih.gov/grants/guide...

grants.nih.gov

NOT-OD-26-131: NIH Operates Under a Continuing Resolution

NIH Funding Opportunities and Notices in the NIH Guide for Grants and Contracts: NIH Operates Under a Continuing Resolution NOT-OD-26-131. NIH

033

Reposted by Jeff Lewis

HittingerLab @hittingerlab.bsky.social · 05/10/2026

As part of my new chair and centre @uofguelph.bsky.social , we are recruiting a new Assistant Professor through the Canada Impact+ Emerging Leaders program: careers.uoguelph.ca/job/Guelph-A...

careers.uoguelph.ca

Assistant Professor (Canada Impact+ Emerging Leader)

065

Jeff Lewis @lewislab.bsky.social · 04/10/2026

I empathize with this, and I think that AI polishing of drafts solves much of the problem compared to AI drafting the entire thing. The AI-generated emails are so generic that it's impossible to stand out. AI polishing the grammar for otherwise original thoughts and writing does not have that issue.

010

Jeff Lewis @lewislab.bsky.social · 04/10/2026

At some level, I empathize with the idea that the amount of time spent on these is not necessarily inversely proportional to how much folks care or are interested in the lab. But the generic template, the 'telling, not showing,' and the circular statements all make it impossible to stand out.

000

Reposted by Jeff Lewis

Michael Baym @baym.lol · 04/10/2026

Ran is cooking with: ‘the only data point I have about this candidate is this: “they had an important task that could determine the rest of their career, but it required a little effort and personal reflection, so they used AI to complete it.”’

27921

Reposted by Jeff Lewis

Rich Meisel @richmeisel.bsky.social · 03/10/2026

We generated chromosome scale assemblies of house flies with different Y chromosomes. You won't believe what happened next!

02611

Reposted by Jeff Lewis

Sasha Gusev @sashagusevposts.bsky.social · 01/10/2026

A searchable mapping between NIH study sections and program officers that doesn't suck. (Claude one-shot) sashagusev.github.io/po-study-sec...

43716

Reposted by Jeff Lewis

Maitreya Dunham @maitreya.bsky.social · 29/09/2026

New paper that includes work from one of my first grad students (Aaron) and one of my most recent (Chris). Experimental evolution really is the gift that keeps on giving. Mutational bias and local genome context constrain adaptive paths in experimental evolution academic.oup.com/genetics/adv...

academic.oup.com

Validate User

1112

Jeff Lewis @lewislab.bsky.social · 29/09/2026

Right. Could still be a coincidence, and hopefully there will be a transparent audit.

010

Jeff Lewis @lewislab.bsky.social · 29/09/2026

All of this is to say that maybe this will be a false start for AI hypothesis generation, but I do think it will be similarly disruptive as genomics (where it also seems from my perspective to hold the most promise).

000

Jeff Lewis @lewislab.bsky.social · 29/09/2026

I do think that AI has a ton of power to generate hypotheses at a much faster pace than was previously possible, much like genomics did 20 years ago (that feels crazy to type). I remember folks (myself included) thinking genomics was somewhat overhyped, and it was for some things, but not others.

100

Jeff Lewis @lewislab.bsky.social · 29/09/2026

Claude identified other possible leads to follow, and this one happened to win out in head-to-head competitions. Were the 'losers' also enriched for Claude users working on them? Ideally, transparent forensic analysis of how Claude code identified these targets would help clarify.

100

Jeff Lewis @lewislab.bsky.social · 29/09/2026

if this was biased by knowing where to look via training data, it’s suddenly both less impressive and raises questions about AI ‘scooping’ and IP theft. With lots of labs looking at different things, especially in this space, it could just be a coincidence that a Claude user was working on this.

220

Jeff Lewis @lewislab.bsky.social · 29/09/2026

So, I've been thinking about the Claude RT preprint/press release. I'm not exactly sure what to think yet without more information. If this really was novel hypothesis generation via large-scale unbiased genome analysis, I think that's fairly impressive, even if premature without follow-up. But,

141

Jeff Lewis @lewislab.bsky.social · 28/09/2026

Raising the storytelling bar with that cool animation.

150

Reposted by Jeff Lewis

Will Ratcliff @wcratcliff.bsky.social · 28/09/2026

New paper out with @kaylastoy.bsky.social: academic.oup.com/evlett/advan...

812458

Jeff Lewis @lewislab.bsky.social · 28/09/2026

I remember this coming up before in the context of that qed preprint rater, and it's not clear if there's any enforcement or recourse.

100

Reposted by Jeff Lewis

Craig Kaplan @triggerloop.bsky.social · 27/09/2026

The RT was annotated in the sequence previously for sure, but what was discovered was the associated clustered repeats that are expressed and appear to associate with the RT, i.e a potential anti-viral system of unknown function but potentially programmable in some way

3494

Jeff Lewis @lewislab.bsky.social · 27/09/2026

We had the heated lids when I started as a grad student in 2001. Never did PCR as an undergrad though.

010

Jeff Lewis @lewislab.bsky.social · 27/09/2026

Pretty sure that I'm older than you, but we had a heated lid for our Perkin Elmer cyclers in grad school. I remember getting MJ ones with a gradient, and those were game changing.

150

Reposted by Jeff Lewis

BWJones @bwjones.bsky.social · 25/09/2026

Science peeps: Looks like SciENcv updated the digital certificate used to sign Biosketch and Current & Pending documents without notification. The old certificate expires Saturday, September 26, 2026. SciENcv docs before that date may fail validation in eRA Commons delaying or blocking submission.

77472

Reposted by Jeff Lewis

Mark Baxter 🏳️‍🌈🧂 @markgbaxter.bsky.social · 24/09/2026

If you work with international personnel, you may want to comment on this proposed rule change that would put visa holders out of status immediately (instead of after a 60 day grace period) when their employment ends keepthe60days.com

keepthe60days.com

Keep the 60 Days: H-1B, E-3, L-1, O-1 & TN grace period

DHS proposes ending the 60-day grace period after a layoff for H-1B, E-3, L-1, O-1 and TN visa holders. Comment on USCIS-2026-0364 before November 10, 2026.

02839

Reposted by Jeff Lewis

Jenna Norton @jenna-m-norton.bsky.social · 24/09/2026

How to destroy strong working relationships, camaraderie and collaboration among staff: Create a system that evaluates staff in comparison to one another rather than based on how well they do their jobs. www.fedweek.com/fedweek/opm-...

fedweek.com

OPM Sets Cap on Top Two Ratings Levels of 40 Percent; Not a "Target"

OPM has set a cap of 40 percent for the top two ratings levels combined in performance evaluations for the large majority of employees under the GS system

910233

Reposted by Jeff Lewis

Harmit Singh Malik @harmitmalik.bsky.social · 24/09/2026

Some random thoughts: Good science doesn't always need disruptors. Often, it needs load-bearing walls. You can't pivot away from a failed multi-year experiment or "iterate" your way out of a mentee's PhD. Science needs stable foundations, not churn. Don't break the people building on top of it.

613022

Jeff Lewis @lewislab.bsky.social · 23/09/2026

That is a very dangerous practice considering AI hallucinations. It's one thing to have an LLM provide references as part of a literature search, that you then go back and read to verify. It's quite another to trust LLM summaries without looking at the primary sources.

020

Jeff Lewis @lewislab.bsky.social · 23/09/2026

Respectfully, raw quantity has always been a poor proxy for scholarship quality. We all know examples of great scientists who publish a small number of large papers and others who publish many more strong, concise papers. I've been beating this drum for evaluative purposes for a long while.

031

Reposted by Jeff Lewis

Drug Monkey @drugmonkey.bsky.social · 22/09/2026

The University of California system contains 6 of the top fifteen public universities in the US, including the top two. This rests on a long-ago vision that public education is a good thing and worth a lot of funding by the taxpayers of California. www.usnews.com/best-college...

usnews.com

34611

Reposted by Jeff Lewis

Dave Baltrus 🦦 @surtlab.bsky.social · 20/09/2026

Ran is correct about this IMO…. There’s use and there’s abuse and there’s laziness with regard to AI and academia. And there’s no one size all answer that works for everything. I do disagree with the footnote but acknowledge old Twitter has been better for science discussions.

3103

Reposted by Jeff Lewis

John Smillie @johnsmillie42.bsky.social · 20/09/2026

Okay, so, Nobel Prize for whomever figures out how to run this horrifying experiment in reverse, where penicillium mold and its cousins are forced to come up with increasingly effective antiobiotics.

1122

Jeff Lewis @lewislab.bsky.social · 17/09/2026

The only way to find typos is by clicking 'submit.'

120

Reposted by Jeff Lewis

Maitreyi Das @daslabpombe.com · 16/09/2026

The Das lab has 1-2 open postdoctoral positions. Our lab investigates the fundamentals of cell-cycle-dependent cell polarity. Boston College has fantastic resources, including a cutting-edge imaging facility with state-of-the-art microscopes. Please apply here: bc.csod.com/ux/ats/caree...

0713

Reposted by Jeff Lewis

Will Ratcliff @wcratcliff.bsky.social · 15/09/2026

1/17 Submissions are now open for the 2027 Microbial Population Biology GRC! We expect this meeting to be extraordinarily oversubscribed. If you want to come, time is of the essence--apply ASAP! June 27–July 2 • Andover, NH www.grc.org/microbial-po...

310572

Jeff Lewis @lewislab.bsky.social · 15/09/2026

And golden tickets are largely redundant with NSF PO behavior. POs do fund from the bottom third (low-priority bin). And it's not rare for a panel to have no high-priority grants, giving POs wide discretion to choose from the medium-priority bin. No percentiles is de facto golden ticket by POs. 2/2

021

Jeff Lewis @lewislab.bsky.social · 15/09/2026

Sigh, so many problems with this. Assuming success rates at 20% and 10 grants per reviewer, 50% of funded grants would be golden tickets. I think 25-75% of funded grants are likely to be golden tickets, depending on how many reviewers use one. This much power is ripe for quid pro quo abuse. 1/2

152

Reposted by Jeff Lewis

Adam Siepielski @adamsiepielski.bsky.social · 15/09/2026

The Department of Biological Sciences at the University of Arkansas is hiring a TT Assistant Professor Position in Ecology. We have a wonderful group of colleagues. Join us! Apply here: tinyurl.com/yxk5marb

Photograph of the University of Arkansas in Fall.

14032

Reposted by Jeff Lewis

Michael Baym @baym.lol · 13/09/2026

Ten years ago Tami, Roy, and I released a movie we made of antibiotic resistance evolving in real time. I thought maybe a couple thousand people tops would see it. I couldn't have imagined it would be seen millions of times, and be taught from middle school through the graduate level worldwide

23437127

Reposted by Jeff Lewis

Maitreyi Das @daslabpombe.com · 10/09/2026

NIGMS MIRA is officially here! This was one of the longest waits ever, but I'll take it. Also, we have openings for a couple of postdoc positions in cell biology/biochemistry. Details to follow soon.

static.klipy.com

Cookie Monster Dancing

ALT: Cookie Monster Dancing

10599

Jeff Lewis @lewislab.bsky.social · 09/09/2026

The official excuse, but I figured they would quickly get drunk on the cost savings.

121