Politically independent 🇺🇸 🇨🇦
Puget Sound
Joined January 2015
- Tweets122K
- Following4.4K
- Followers8.6K
- Likes18K
"scaling scientific attention" What a brilliant way to describe what LLMs can actually do for wet labs.
A few thoughts on Anthropic’s ART discovery, since it’s close to the work we do:
The biology is intriguing. In phages, an unusual reverse transcriptase sits next to a partner gene and an array of DNA repeats. The team found that the array produces distinct short RNAs.
It’s tempting to compare this with CRISPR or retrons, but we don’t yet know what ART does. Whether the RNAs guide anything, or whether the system is programmable, remains an open question.
I’m just as interested in how they found it. About 950 Claude agent sessions ran over 21 hours, surveyed ~200,000 reverse transcriptases, and produced 19 reports. One agent looked at the DNA next to an RT and spotted a repeat array outside the features the search had set out to examine.
It followed that observation, compared it with known systems, checked the literature, and gave the scientists something concrete to investigate.
To me, that’s the opportunity: scaling scientific attention. We have more biological data than any team can inspect closely. Agents can help us notice the odd cases and show their work.
There’s still reason to be careful. Repeat-rich regions in assembled phage genomes can be tricky, so I’d want to see raw read support across the arrays and their boundaries in independent isolates. And the most important experiment is still ahead: what does ART actually do?
I think discovery will become a tighter loop. Agents propose hypotheses, predictive models help choose experiments, and automated labs generate results that inform the next round.
We’re working toward that at @Xaira_Thera, bringing agentic workflows together with models such as X-Cell and wet lab experiments. I’m excited by how much more biology we may be able to explore when each experiment helps us decide what to test next.
There will be a data center in every crawlspace and basement to heat our homes and replace our lost salaries.
Anyone was has purchased large numbers of ASICs to mine bitcoin has understood for a long time why $NVDA is unstoppable.
The sobering thing about this AI correction is that nothing has fundamentally changed except that a lot of people started talking about it.
My daughter scored 79/150 on the Digits of Pi Quiz piday.org/pi-quiz/
![Pi Quiz, Memorize Pi [Easy And Fun Memory Game] | Pi Day](https://pbs.twimg.com/card_img/2118648749400473600/NPLuFh83?format=jpg&name=800x419)
Nondescript retweeted
Here is another incredible GPT-5 Pro example that is well above biomedical PhD level and how it advances science:
I was reviewing some of our old data and noticed an email from one of my PhD students sent more than eight years ago, outlining a highly complex immune cell experiment that would run for several weeks and asking me to make corrections. I copied and pasted her experimental plan into GPT-5 Pro and asked for suggestions. Its response was significantly better than what I had sent my student back then!
Not only did it figure out the exact purpose of the experiment and make all the corrections I had made in extreme detail, but it also suggested a couple of changes that would have made the output easier to interpret and made the experiment better, which I had unfortunately missed. It also created a detailed spreadsheet with a 96-well plate layout for cell seeding and different conditions. We usually design these to avoid errors, and this one was perfectly organized to make setup easier.
Then I asked GPT-5 Pro to run various outcome scenarios, interpret the results accordingly, and propose new experiments. Again, its interpretations and suggestions were simply outstanding, including a couple of ideas we had missed!
Incredibly, GPT-5 Pro would have been as good as, if not better than, me at making these corrections, interpretations, analyses, and follow-up experiment suggestions! The experiment would also have yielded better results thanks to more precise planning, and that foresight would have enabled quicker planning of the next experiments!
I don’t care what people say about the limits of current AI models or their nonsensical criticisms; they have no freaking idea how transformative its impact on the scientific process will be! It will advance and accelerate science far beyond what anyone is imagining and make the world so much better place!
Why the hell did Amazon remove the ability to archive orders - do they not understand the idea of gift giving, surprise, and standard traditional practices. What kind of autist crap is this?
