Sasha Pyshkov

ML practitioner, AI enthusiast, space geek, Vive le Canada

[Sasha Pyshkov @Pyshkov]

I have a tendency to ask every new model to do something unusual in R (which it was not really built for)... I asked Claude Opus 4.6 to make a "movie" in R code.. Here's the result, and I am mindblown.

[Oliver Venture @_OliverVenture]

$BEAM

"Beam Therapeutics Announces Clearance of Investigational New Drug Application for BEAM-304 for the Treatment of Phenylketonuria (PKU) by the United States (U.S.) Food and Drug Administration"

[Patrick Hsu @pdhsu]

"Less than a dozen of these machines operating at full speed can do more full body scans than every MRI on earth" 🤯

[Midjourney @midjourney]

Announcing a new division of Midjourney called "Midjourney Medical"

[Kydo @0xkydo]

🤯 Midjourney -- yes, the AI image company -- just shipped a brand new type of imaging machine. 🤯

- 100x faster than an MRI.
- 10x cheaper.

Full body scanned in 60 seconds instead of an hour in a tube. Ultrasound based, MRI-level resolution.

And it's real -- not a concept, a working machine. You step into a shallow pool of warm water, a ring of half a million sensors sends sound through your body from every angle, and ~60 seconds later you have a 3D map of your insides down to a fraction of a millimeter. No radiation, no tube, no lying still.

They're not even building it as a hospital machine -- they're building a spa. The scan is a side-effect of a place you'd want to hang out anyway.

Lastly, it is built by 9 people. NINE PEOPLE.

You can just do things.

[Klaus Molle @KlausMolle]

$ntla rich resource on KOL & investigator Dr Banerji’s perception of supreme HAELO phase 3 results for lonvo-z in HAE:

Patients in the trial reported being able to think differently about work, school, and travel

https://t.co/aPsbqx1qHm

[Ethan Mollick @emollick]

I have a fun, oddly useful AI benchmark: "build me a procedurally generated 3D simulation showing the evolution of a harbor town from 3000 BC to 3000 AD, it should look beautiful & allow me to have some control over it"

Play the gallery of 20 models: https://t.co/zN2uHY1gl8

[Wei Zhao @zhaoweiasu]

$BEAM risto-cel (BEAM-101) EHA 2026 update
➡️ Total 44 patients dosed
➡️ No sVOCs
➡️ All patients achieved rapid and robust increases in total hemoglobin (Hb) and HbF (>60%)
➡️ At M6 (n=29), the mean endogenous HbF was 63.5% and HbS was 36.2%
➡️ HbF+ cells reached >99% at M2 and remained >99%

[Sasha Pyshkov @Pyshkov]

@CaptainRisky7 @GeneInvesting Indeed.. more fun ahead

[Gene Investing w/Anthony 🧬 @GeneInvesting]

Love waking up to headlines like this.

$PRME

[Gene Investing w/Anthony 🧬 @GeneInvesting]

Good morning to those who believe in the power of curative gene editing medicine ☀️ $NTLA

[Avi Roy @agingroy]

Imagine fixing a genetic disease with one IV bag.

Hereditary angioedema causes sudden, suffocating swelling attacks. The new treatment infuses CRISPR straight into the bloodstream, where it edits the broken gene inside you. No lab, no harvested cells.

In a Phase 3 trial of 80 people, one dose cut monthly attacks by 87%, with no serious side effects in the treated group.

One infusion. One edit. A lifelong disease, quieted.

[Nick Mark MD @nickmmark]

Pretty exciting use of CRISPR!

Hopefully this disease - and the challenging airways it leads to - becomes a thing of the past!

[Eric Topol @EricTopol]

A life-threatening inherited disease, hereditary angioedema, with striking benefit from one-shot CRISPR genome editing.
Phase 3 randomized, double-blind trial results @NEJM today
https://t.co/pT4fVgZcNP

[Dr. Brian Goldman @NightShiftMD]

I’ve seen several patients lately with life threatening angioedema. It looks like anaphylaxis but behaves differently and does not always respond to an EpiPen.

[NEJM @NEJM]

Presented at #EAACICongress2026:

In a phase 3, randomized trial of lonvo-z, an in vivo gene editing therapy, patients with hereditary angioedema who received the drug had a significantly lower attack rate than those who received placebo. Full HAELO trial results: https://t.co/qBliFvKFKl

@EAACI_HQ

[Sasha Pyshkov @Pyshkov]

@KelsiIsaiah 100% agree, I am excited for HAE patients - but even more so for the near future gene editing revolution

[Kelsi Isaiah @KelsiIsaiah]

$NTLA I understand the attack reductions are the more fundamentally important measures of these trials. But take a look at the AE-Quality of Life Score - it’s almost 4x the clinically important improvement.

Thank you NTLA, this is literally life-changing for AE patients!

[Unvested Investor @Alex___VA]

$NTLA we're watching the transition from chronic disease management to functional cures happen in real time. The HAE data from Istanbul was remarkable.
The ATTR setback will be remembered as a temporary detour driven by an exceptionally complex patient case, not a platform issue.

[3PrimeAnalytica @3primeAnalytica]

$NTLA great data, universal benefit, zero SAEs + simultaneous @NEJM publication...lfg!

→ huge QoL delta, (AE-QoL) total score change from baseline to Week 28: −23.51 vs −6.47 pbo (treatment difference −17.04....a ≥6-point drop is clinically meaningful....this is nearly 3× that!! Every patient remained off LTP

→ fully confirmed, all pts in the active arm showing reductions from their own baseline.

→ key secondary: attack-free AND completely off all other HAE therapy for the full 6mo period.

→ Moderate-to-severe attacks: 91% reduction (0.11 vs 1.23).

→ kallikrein dropped rapidly (by first measurement ~day 15), reached nadir by week 5, and stayed stably suppressed

→ Attack rates remained well below pre-screening SOC levels in longer follow-up.

→ zero SAEs and clean labs in the active arm.

As I called in April, the 62% is conservative and real-world will look even better!

[Dirk Haussecker @RNAiAnalyst]

$ntla the kalleikrein balancing act was successful. Was this the endgame in HAE?

I doubt anybody, except for maybe $ions, will start investimg in a next-gen HAE prevention medicine as long as ntla-2002 is on the market.

Commercialization will be a success and give entire genome editing industry a big lift. Free alpha for y'all.