Rory Kirchner

[Michael 英泉 Eisen @mbeisen]

Whenever @lpachter reads a bioinformatics paper

[Rory Kirchner @RoryKirchner]

@VirusesImmunity @TrendsImmuno @jannamoen @ChrisMillerDO is her Twitter ID!

[Prof. Akiko Iwasaki @VirusesImmunity]

Our review is highlighted by the cover of the current issue of @TrendsImmuno -The lingering shadow of epidemics: post-acute sequelae across history by Christine Miller and @jannamoen.
Please read (open access)👇🏼
https://t.co/gCS7nmVKUW

[Luca Pinello @lucapinello]

1/ Thrilled to share our DNA-Diffusion paper, co-led by @bioinfolucas & @srsenan , is now out in @NatureGenet! New experimental results: STARR-Seq validation across 3 cell types (episomal) and EXTRA-Seq for endogenous activation of AXIN2, a leukemia protective gene!

[Jacob Schreiber @jmschreiber91]

In our initial preprint, we showed that we could use Ledidi to design regulatory DNA in silico and validated the designs using external models.

In this preprint, we have designed actual enhancers and experimentally verified them.

https://t.co/Afh9AUDgzG

[Cell Press @CellPressNews]

To effectively treat long #COVID, we must learn from historical chronic illnesses, medical researchers say. https://t.co/f2MhdA322x

Akiko Iwasaki, Christine Miller, Janna Moen
@TrendsImmuno

[Rory Kirchner @RoryKirchner]

@VirusesImmunity @jannamoen @ChrisMillerDO

[Prof. Akiko Iwasaki @VirusesImmunity]

Check out our new review, "The lingering shadow of epidemics: post-acute sequelae across history" by Drs. Christine Miller and @jannamoen. We highlight historical accounts of post-acute infection syndromes (PAIS) through the centuries 👇🏼 #LongCOVID
https://t.co/gCS7nmVKUW

[Surag Nair @suragnair]

Excited to share Nona: a unifying multimodal masking framework for functional genomics.

Models for DNA have evolved along separate paths: sequence-to-function (AlphaGenome), language models (Evo2), and generative models (DDSM).

Can these be unified under a single paradigm? 1/15

[Rory Kirchner @RoryKirchner]

@EmilSutovsky Yeah all of us non close friends with Danyas phone number should have just texted him support is what you meant. Right. People with good character should be leaders. That’s not you. You can’t even apologize correctly. Resign.

[Rory Kirchner @RoryKirchner]

@vsbuffalo Baseball is the best game to have on when you sr doing other things because it has very simple states so you can only partially pay attention and also catch everything. Perfect radio game.

[Rory Kirchner @RoryKirchner]

@owl_posting I don’t read very many blogs but I always read yours.

[owl @owl_posting]

How do you use a virtual cell to do something actually useful? (2/3)

In this second case study, we explore how our research at @NOETIK_ai suggests better inclusion criteria for a real-world, 2024 cancer clinical trial. Link below!

As we mentioned in our last essay, the inclusion criteria for many cancer therapies are shockingly simplistic. They routinely exclude patients who might benefit, while enrolling others who have almost no chance at response. Oncologists are not unaware of this complexity. The real barrier is that designing new biomarkers is nearly impossible under today's system. It requires years of planning, the right samples, and the foresight to measure the exact markers in advance. The result is grim: almost no new predictive cancer biomarkers have reached the clinic in the last 30 years.

Virtual cells change the starting point. By training a model on an extremely high-fidelity atlas of human tumor heterogeneity, we can explore cancer biology in silico, ideating on far better inclusion criteria entirely through the models embedding space. The essay contains one example of this, but the framework is extremely generalizable, and we are already extending it across other therapeutic areas.

If you're working on trial design, biomarkers, or translational oncology and want to explore what's possible, we'd love to hear from you! Reach out at [email protected]

Link: https://t.co/N37pOZxc8N

In the next, and final, case study coming next Friday, we'll finally turn to the prototypical utility of virtual cells: perturbations.

[Richard McElreath 🐈‍⬛ @rlmcelreath]

For the new kids in back: If you hate statistics, you'll love my free lectures. Putting science before statistics, from basics of inference & causal modeling to multilevel models & dynamic state space models. It's all free, made with love and sympathy. https://t.co/GnOYGex9Yg

[Prachee Avasthi @PracheeAC]

Started a substack with my favorite person to argue and solve problems with, @mbeisen. If you love science as much as we do and want to see it reach its potential, come follow along and join the conversation

Our first post below 👇

[Rory Kirchner @RoryKirchner]

@vsbuffalo @nathanfielder I have the same hobby— seeing how all of the holes in the Swiss cheese line up to make a disaster happen is instructional. I have a related thing you may also enjoy. https://t.co/TijoMakBT3

[Rory Kirchner @RoryKirchner]

@vsbuffalo Six months ago I would have taken you up on that bet. Now, no. They got a lot better than I thought was possible in the last six months.

[Rory Kirchner @RoryKirchner]

@SenSchumer I am done with you cowards. You are not leaders.

[Rory Kirchner @RoryKirchner]

@vsbuffalo There is actually incentive to do poor quality work, as long as it isn’t easily catchable by a reviewer— you’re more likely to get a strong result if you are sloppy, which is more likely to get noticed and published in a top tier journal. It’s also quicker to be sloppy.

[Rory Kirchner @RoryKirchner]

@NateSilver538 Changing your mind in response to new information doesn’t mean you were previously lying.