Michał Jastrzębski

Spaghetti is what we call sections of protein that aren't structured and are generally floppy. When you think about it, proteins themselves are, quintessentially, pasta.

Did you consider steam deck as a platform to control lab stuff? Lots of compute, great battery, all the inputs you want, Linux on board and it can run Skyrim!

As long as it's Japanese 7-11, where onigiri is 10/10.

If you have played pillars games, it's amazing. Mostly because it fits so well into this world building. If not, I guess it won't hit nearly as much. Strongly recommend PoE games before avowed

Same here, I've already ordered the disk. It'll take a while to download the genbank in it's entirety

Nice thing about torrent approach is that nobody (besides the first person) has to host the entire thing, as long as everything is available from someone. If you can only spare 2tb, just pick least available 2tb subset and host that

I actually disagree. I believe there's enough people supporting science for this to happen. Folding at home works to this day and it's much more expensive to run than seeding a torrent

There's filecoin, it would do that, but I'm still for hoisting the colors and sail torrential seas

No need. Torrent networks show that people will share data. If this works for pirated movies and porn, I kind of hope it'll also work for science

I was just looking at this. Let's make genbank torrent. I'll seed forever

Hey fellow bio kids! Does anyone know cheap, diy and scrappy cell lysis buffer? Something that'll melt yeast and other common bugs. Trying to get into metagenomics

It’s been a tough few weeks. My 10yo daughter was diagnosed with a very rare, aggressive cancer called interdigitating dendritic cell sarcoma (IDCS). I’m reaching out to identify clinicians/patients who have encountered pediatric IDCS or other (non-LCH) dendritic or histiocytic sarcomas cases.

Predicting RNA-seq coverage from DNA sequence as a unifying model of gene regulation - Nature Genetics Borzoi adapts the Enformer sequence-to-expression model to directly predict RNA-seq coverage, enabling the in-silico analysis of variant effects across multiple layers of gene regulation.

Super excited to announce our latest flagship model Borzoi: major props to Johannes & David Kelley et al for advancing it. It's been a long journey from our prior Enformer model into this one. A few innovations: i) longer DNA context, ii) adaptation to predict RNA-seq abundance and splice isoforms,

GitHub - ginkgobioworks/ginkgoai-protein-embedding-tutorial: Tutorial on how to use Ginkgo AI embedding APIs for scientific problems Tutorial on how to use Ginkgo AI embedding APIs for scientific problems - ginkgobioworks/ginkgoai-protein-embedding-tutorial

Hey! I wrote a thing! github.com/ginkgobiowor... This is a tutorial on how to use deep learning to solve scientific problems without being an ML scientist. Use the Ginkgo API to predict all sorts of things. Lmk what you think

bsky.app/profile/stil... this one started my chain of recommendations

I got swamped by all the bio science starter packs and now my feed is all nerdy. Lovin it

RIP Netflix servers

Keep calm and build useful models. The only way to cut through the noise

This is 100% true. I’ve been seeing this for a while. Knowing how things work deeply is a super power, and one you can use to create better products.

Can I join the cool kids club?

Well, they took high interest safety loan by tying their future to Russia (as a Pole, it's definitely a deja vu). Now it's due time and lots of things they took for granted disappeared.

Elitality

Revolutionizing large-scale DNA synthesis with microchip-based massive in parallel synthesis system DNA synthesis serves as the fundamental enabling technology of engineering biology, aiming to provide DNA molecules of designed composition, length, and complexity at scale and low cost. Current high-...

If this holds up, then we're so back www.biorxiv.org/content/10.1...

My gradients? Exploding

I can think of several setups I've deployed that fits

As someone who spent the last 10 years of my career doing machine learning, I want this T-shirt

Step 1. Start bot net screaming something like "I bet on Trump MAGA!11!oneoneeleven". 2. Bet on Kamala. 3. Gather the pot

Polymarket, in particular, is interesting here. There's financial benefit for some users about it being wrong. You'll get more money if polymarket thinks Kamala loses, and you bet on her. I'm curious if it's 9000iq mind games trying to convince more people to bet on Trump for bigger payout

Did you check affinity photo? I'm very happy with it so far, hadn't go back to Lightroom since

To me it's not either or. We should do all we can. Biomanufacturing, clean power, emission reduction, all of it. My point is, like VCs or not, they have money and they can put it into good or bad. If e/acc stan funds clean power or synbio company for tweet clout, it's still better than crypto.