Robert Y. Chen
@therealRYC
PGY-3 @UW Psychiatry. Prev MD-PhD @WUSTL. Biomarker and drug discovery for psychiatry, starting with Schizophrenia. Dog Daddy x2 to Taro and Azuki.
This is actually bananas. If any treatment in medicine could reduce risk of physical trauma by 7x, it would be approved immediately. And, FSD will only get better with time. Your and your family’s safety does not care about people’s opinions of @elonmusk or @Tesla 🤯
NEWS: Tesla has revealed that in Q2 2025, they recorded one crash for every 6.69 million miles driven in which drivers were using Autopilot technology. For drivers who were not using Autopilot technology, @Tesla recorded one crash for every 963,000 miles driven. By comparison,…
Certainly not helping trust in scientific institutions 😔 “By 2024, the number of papers using FinnGen data grew by nearly 15 times from 2021, for example, while those using FAERS increased by nearly 4 times and UK Biobank by 2.4 times over the same period.”
Data from five large open-access health databases are being used to generate thousands of poor-quality, formulaic papers go.nature.com/4lz1Zch
There were quite a few interesting papers on genetics, omics, and deep phenotyping last week👇
Generally good advice for all fields of inquiry. When conflicting sources of evidence exist, be open to dropping your "darling" view and come up with new hypotheses that are consistent with all available data.
Knowledge in medicine and clinical psychology requires a careful triangulation of information obtained via multiple modes of inquiry: clinical observation, clinical trials, observational studies, patient reports, and basic research. The most robust conclusions show a strong…
Why “reasoning” goes beyond just typing out sentences that sounds like how humans reason. Perfect prediction of planetary orbit with 0 understanding of gravity. AI progress does not mean blindly trust AI, it means humility and careful optimism
Can an AI model predict perfectly and still have a terrible world model? What would that even mean? Our new ICML paper formalizes these questions One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws 🧵
Do you know what is the baseline rate of congenital malformations? ( non drug exposure) And do you know what is the incremental rate with SSRIs? Do you know how it compares to 1. Depression 2. Diabetes 3. Maternal obesity? Did you even attempt to clarify or provide this…
If serotonin plays a crucial role in fetal development (which it does) & if SSRIs disrupt the serotonin system (which they do) then the SSRIs must alter fetal development - and the public should be warned. This point was not disputed at yesterday's panel.
This week in Psychiatry 🤯 🧬How important is genetics in Psychiatry? 😴Inflammation and depression, but only in insomnia 🕵️Different psychosis subtypes, backed by genetics 🥷SeaportTx's 'Stealth Mode' Antidepressant ❌FDA gives thumbs DOWN for PTSD treatment Bookmark, share,…
x.com/i/article/1944…
Hi David, don’t disagree with your core message. However, I do want to correct the record that @formationbio always planned to be a full stack Pharma company, differentiated by faster/more efficient clinical trials + drug development. In fact, this was articulated in our seed…
Really incredible story of how misdiagnosis can rob you of a fulfilled life. I have many soapboxes in psychiatry but the biggest one is spending the time to give the right diagnosis.
Mental illness cost Imtiaz Zafar almost a decade of his academic life. Now it motivates him to help find effective treatments go.nature.com/46JIpVW
Probably the best objective take on the ongoing discourse on antidepressant discontinuation syndrome. Knowledge presented with nuance is a winning formula.
🧵8–12 Week Trials? The Real Problem in Antidepressant Withdrawal debate Is the Dichotomy, Not the Duration 🚨1/10 Many critics argue that trials on antidepressant discontinuation are invalid because they’re based on only 8-12 weeks of use. But here’s the paradox: If…
Introducing Happy Warriors -- stories of founders building what civilization needs. Episode 1 is here, featuring @Plasmidsaurus, the startup quietly speeding up biology worldwide.
“The money tells the real story. @pfizer just paid $1.2 BILLION upfront—not total, upfront—for one Chinese cancer drug. 2015: China got 0% of global pharma deal money 2025: 26% and climbing When @Merck loses sleep over a drug from Akeso (who?), you know the game has changed.”
Everyone missed the real China story. While we debated TikTok, Chinese scientists quietly built 1,250 novel drug candidates—up from 160 in 2015. That's not a typo. They added 1,090 innovative medicines in 9 years. The entire EU added 200. Here's how they broke pharma's…
I don’t think the piece paints @sameerjauhar perspective accurately. In the discussion of their paper, they acknowledge the fact that the study used mostly short-term trials. They cite many self-limitations. I do think the messaging of the article is correct though - that for…
"If read uncritically, this new study could cause significant harm by misleading patients, clinicians, guidelines and policy makers." Plain speaking piece from Simon Opher, Labour MP and GP bit.ly/4kDAfSw
Knowledge in medicine and clinical psychology requires a careful triangulation of information obtained via multiple modes of inquiry: clinical observation, clinical trials, observational studies, patient reports, and basic research. The most robust conclusions show a strong…
Pretty interesting trend where Chinese biotech can breeze through Phase I/II enrollment, but are only on-par with USA in Phase III recruitment. I wonder if this is due to funding limitations where smaller-stage biotech companies are capital-strapped and have slower recruitment…
Nice visual representation of China's current advantage in patient recruitment / trial enrollment Oncology & Obesity Ph1s & 2s enroll 6-9 months faster in China than in the US H/t @business
I personally think from a drug discovery perspective that it will still be 10-15 years before we “validate” whether AF3 and other models optimize the clinically useful endpoint of Predicting Phase 3 success from molecule structure. Anything upstream of that could be really…
I have a question for those who believe many biology problems are AI-ready, similar to static protein structure prediction. What are the major results since AlphaFold was introduced?
This week in Psychiatry 🤯 ❌ Raging debate on antidepressant withdrawal 🧩 Redefining autism spectrum disorder 🥷 Draig Therapeutics emerges from stealth ($150M) 💡 MapLight's M1/M4 modulator enters the clinic Bookmark, share, and let me know your thoughts below!
x.com/i/article/1943…