Профиль

pretty good roadmap showing where to apply AI to speed up science

We don’t give the Trump administration enough blame for a disastrous COVID response.

Pretty sure this is illegal? certainly bad faith

Jumper to Anthropic! DeepMind putting out AI scientist papers!

Incredible alternate DNA defense hacking. Bacteria uses sulfur instead of oxygen in its DNA, making it resistant to its own anti-phage nucleases. Phage says gimme some o‘ that

i know this is my particular lil niche but i think it's prettttty interesting!

blowing up nord stream was the largest single human-caused release of methane in history. all respect to ukraine and their righteous fight, but the war is an environmental catastrophe.

thank god they came back negative. my scary result was a false positive. defect in my design. so i got back to work. but that, children, is why you shouldn't do experiments on yourself. and if you must, at least include the proper controls!

i had done this many times before, so i didn't include it with that day's samples and man was i kicking myself for not running it. my hands were freaking shaking doing the pipetting, then the longest two hours of my life waiting to get the results.

by now i'm sweating and dreading the shit storm this is gonna stir up and the potential to spread it to some key people who we really didn't want to get sick. but we decided to run my samples again using a standard primer set for comparison.

i've been in and out of the hospital all day hauling shit between two labs. we're under strict masking and hand washing protocols but this is not looking good. my lab mate, part of the testing team, comes in and i'm like hey i got some bad news can you take a look at this? he agreed it looked bad

the picture flashes up on the screen and horror of horrors my sample looks exactly like the one with viral RNA added. big fat PCR band right at the expected size. and my control without my swab sample is completely blank. and i'm like "ooohhhh shit" then "what do i do? do i just leave?"

so i'm in the lab. alone of course because everyone was ordered home. and i'm trying out a new primer pair that i designed to PCR amplify a specific region of the virus. I read out the PCR products by size on an agarose gel, and i get the results by taking a photo in a special illumination chamber.

we were working with Procter & Gamble to injection mold plastic replacements for the Copan swabs. little gray sticks with funky shapes on the end to grab human cells from (my) nasal swabs. eventually the US gov't made a deal with US Cotton to mass manufacture cotton swabs. you may have used one

actually i was sticking a variety of foreign objects up my nose. Milan, Italy was the world's single supply point for the gold standard "Copan swabs", and it had been hit with a massive COVID outbreak, the first outside China IIRC, so everyone was hustling to find and validate replacements.

supply chains were totally fucked at the time and we couldn't get standard laboratory controls. mucus with a realistic amount of human DNA is important background "matrix" for a diagnostic test, so I was giving myself the old brain tickle on the regular. for science

reminds me of the time i diagnosed myself with COVID. i was developing experimental high volume COVID testing methods. every morning i gave myself a nasopharyngeal swab as my negative control, then spiked in SARS-CoV-2 RNA for my positive sample.

Hate to say effervescent when I mean evanescent

Feel like the proportions matter and the buildout is mostly for training and hosting LLMs. Data centers are local politics. individual preferences carry weight. The data center backlash is an AI and big tech backlash focused on the one place where government is responsive

cool! RNA mutagenesis pulse-chase to measure RNA dynamics by sequencing. looks great