Jim

[Miles Cranmer @MilesCranmer]

This is an insane paper and I love it

https://t.co/DP8OR5NJf2

[Jim @jimtalksdata]

Blocking single proteins never really made sense to me, frankly, with the combinatorial explosion that PPIs get you (an N^2 process vs a N process). Much more specificity using PPIs

Funnily enough imatinib was also based on "PPI" (though in that case the proteins are fused)

[Jim @jimtalksdata]

Devilishly clever.

"In the end, it was protein-protein interactions all along" (topic of my grad thesis)

[Parmita Mishra @parmita]

two flavors, broadly:

PROTACs
pull the target to the cell's garbage disposal and have it shredded.

molecular glues
force two proteins to stick that never would, building a brand-new function.

suddenly "undruggable" just meant "we hadn't found its partner yet."

[Jim @jimtalksdata]

The imatinib of this generation.

[Mark Lewis, MD, FASCO @marklewismd]

Cheers, chills, and a standing ovation when RASolute 302 showed unprecedented survival on daraxonrasib for patients with progressive pancreatic cancer

Seldom do you sense you’re witnessing a historic moment in cancer care but this feels like ras targeting has arrived

#ASCO26

[Parmita Mishra @parmita]

RAS finally getting drugged is one of the great stories in modern biology, and almost nobody outside oncology understands why it's such a big deal.

YOU'LL LEARN SOMETHING AWESOME TODAY.

i am going to keep this as understandable (and simple) as i can.

OPEN THE THREAD.

🧵

[Jim @jimtalksdata]

@smlungpathguy As molpath, it’s fun to “rewrite” these entities in terms of molecular alterations:

CDH1-deficient tumors:
ILC and diffuse gastric ca (classic)
Urothelial ca, plasmacytoid variant
High grade colorectal (rarely)
Gleason 5 prostate ca (also rarely)
Others

https://t.co/GTilCMwzOD

[Jim @jimtalksdata]

@commonsense_md @Jkylebass @POTUS @SecRubio @SDNYnews @SecScottBessent That is the question US college grads are asking. Address that, and the number of qualified people goes way up.

[Jim @jimtalksdata]

@commonsense_md @Jkylebass @POTUS @SecRubio @SDNYnews @SecScottBessent Recent grad. AMGs (US MD/DO) have a massive advantage over IMGs/FMGs in terms of residency selection; the barriers for FMGs to successfully practice in US are massive. However, foreign medical schools are much less costly. Would you take 500k in debt and uncertainty to be an AMG?

[Jim @jimtalksdata]

@awealthofcs @D8vidSmith I think using a VC exit strategy allows for sensible sizing.

Market cycle - “round”
Multiples (exit vs invested) - 10-40x (depends on risk tolerance). Aggregated, so actual appreciation might exceed that
Tail/residual - your rough estimate of store of value / global assets

[Jim @jimtalksdata]

@olsonplanner Traditionally path has had to do a fellowship to find jobs - however that is changing due to the job market and retirement cliff.

Nevertheless it may make sense for compensation (derm/hemepath), work life balance (molpath positions are largely hybrid/remote), or other reasons

[Jim @jimtalksdata]

@olsonplanner 3. Benefits - benefits for insurance etc can make the difference, especially if the spouse is in some high-paying PP/1099 job or small business.

[Jim @jimtalksdata]

@olsonplanner 2. Workload/hours - anecdotally, it is more chill. Some use this opportunity to supplement with locums/consulting work which juices up the returns.

[Jim @jimtalksdata]

Do benchmarks matter?

https://t.co/51qxHNJyup

[Jim @jimtalksdata]

Interesting that the btc complex is acting as risk-off today, despite this clearly being a liquidity event. Won't be too quick to judge though.

Dynamic correlations are dynamic

[Jim @jimtalksdata]

@LynAldenContact @BangBitcoin *currently*, btc is correlated with general risk on bets. I suspect that will change as noise traders make up a proportionally smaller piece of market.

[Jim @jimtalksdata]

@smlungpathguy There also

“Clinical correlation highly recommended/required” - did you send us the right specimen?

“can be considered” - hmm… interesting

“Is necessary to interpret” - did you forget biomarkers for this breast?

😂

[Jim @jimtalksdata]

@smlungpathguy @grok @AskPerplexity Autoregressive models also important here, as most path works that way. Eg clinical/radiology findings? what is your initial impression at low power? High power? IHCs? Special stains? Should we order molecular? The information for future steps depends on the previous.

[Jim @jimtalksdata]

@smlungpathguy @grok @AskPerplexity Or bring multimodal models that will basically measure the sizes and distributions of all cellular/subcellular features, and make a statistically grounded prediction based on that.

We’re not quite there though, though researchers/companies are working on it.

[LaurieWired @lauriewired]

The code is published online, just a tiny bit of Python.

Here’s a breakdown of the regex one-liner:

- lookbehind to skip signal peptides
- scan for rare amino pairs (KK, KR, etc)
- prevent false cleavage match
- enforce minimum spacing + recursive pattern match