Siyu Chen 陈思羽 (@schen_sc) profile picture

Postdoc with Dr. Goodarzi @arcinstitute Former UCSF, McGill and IGDB

Joined October 2021
Photo by Siyu Chen 陈思羽 (@schen_sc)
pretty interesting
🦍 Introducing Memorilla: memory as a modality for LLMs Memorilla turns everything a frozen LLM needs to remember into just 16 memory embeddings ⚡ It uses up to 4,212× fewer tokens and is 27× faster than full context inference while more accurate 🎯 Pushes the Pareto frontier over RAG and agentic memory in personalization, long-document and knowledge QA A thread 🧵:
3
2
3
372
Siyu Chen 陈思羽 retweeted
New preprint on generative modeling for mouse development! Rather than single cell trajectories, we set out to model embryos as trees. Led by @gokl_g, @njwfish, @WilliamNColgan, w/ Philippe Rigollet, @stats_stephen, Zack Smith, @marinkazitnik, @lukekoblan and a wonderful team
3
30
7
151
20,539
Siyu Chen 陈思羽 retweeted
Replying to @AnthropicAI
@AnthropicAI life sciences seems to be doing great work so far. Two questions folks are asking me: Will Anthropic discover cures? Absolutely. For many diseases, staying uncured is a biotech financing failure, not a science/technology failure. Many rare diseases are mechanistically understood and have established genetic targets, and what they are stuck on is market failure. Ant is shaking up the field by throwing VC money at a problem the scientists who study these diseases could never unlock on their own. A startup trying to do what Ant is doing with the same amount of capital simply can't exist. So the better question is: will their AI help, or is it all just capital? Of course the AI helps... It is already helping us scientists do our jobs better, and every rare disease program it makes cheaper is one more that becomes fundable. Do I wish I lived in a world where I could cure a rare disease we have studied in our lab and see it find its way into patients? Absolutely! We have tried (and failed) to fundraise in the past around rare diseases... and even now our efforts in this area (like @thernabio's N of 1 programs) require non-market reasons to exist. But of all the things Ant could spend its effort and capital on, this is one of the most worthwhile I can think of.
7
11
90
5,824
Siyu Chen 陈思羽 retweeted
Surprise 1: Anthropic acquired a one-person biotech . . . Surprise 2: that person is me Surprise 3: I’m here to tell you all about it Yes, this is what you think it is; after 6 yrs of increasingly loose anonymity: the Professor Oak reveal post …and yes, it has gotten so boring around here that I’m literally doxxing myself to spice up the timeline ------- Ok, allow me to introduce myself. My name is Cody Tranbarger. I mostly tweet, but occasionally find time to dabble in drug dev & starting/building biotechs. My past lives span the VC, crossover, startup continuum; Prof Oak has been my way of keeping a digital foot in those worlds, despite having my physical feet planted elsewhere. The past 18 mo, ‘elsewhere’ was Mithril Medicine, a co. I founded to develop bsAbs for LAMA2-CMD & other ECM-related genetic diseases. Mithril was born out of hands-on building with @MikeNGladstone, @JasonPaulRhodes, @LifeSciVC, & team @atlasventure. Somehow, I made enough half-decent calls along the way (even a broken clock is right twice a day, eh?) to convince them Mithril/I warranted a small seed. While I can hardly rationalize such an outrageously cavalier bet, I’m grateful for it, & I'd partner with Atlas again in a heartbeat. In typical AV fashion, Mithril’s primordial form was virtual, decentralized, & ultra-lean. This stage tends to be short-lived, but for Mithril, it worked surprisingly well, for surprisingly long, thanks in part to the rapid maturation of the AI tools I used liberally. Eventually, Mithril became a bit of an experiment itself - to find the upper bound of how far it could scale as a single operator + a bunch of Claude (+ a stellar bench of part-time consultants). In parallel, I started spending time with @nc_frey, @an1lam, & @CoefficientBio team; and promptly experienced, for the 1st time, AI at the frontier. Soon thereafter, Mithril stopped looking like an experiment, & started looking like a preview of an entirely new operating model. From that spark came “Pipeline-in-a-Person.” The idea immediately resonated when we shared it here back in April; and it was central to the shared vision that led to Anthropic’s acquisition of Mithril, & my joining the team. Why join Anthropic? Three reasons. (1) For Mithril, the programs continue, with more resources & infrastructure behind them, and as much Claude as I can put to work. (2) For me, the chance to start several more programs, & find out how many one person can actually run in parallel - an almost unprecedented luxury for a curious drug hunter. (3) For our mission, a path to institutionalize & scale this new way of working, then put it in the hands of all of biopharma, all over the world - which, candidly, I find the most energizing. Will AI x Bio cure diseases? Unequivocally yes, imo… but not by the route that gets the most airtime. Rather than timeline acceleration & compression, I see orchestration & parallelization as today's low-hanging fruit. Most drug hunters see a handful of go/no-go calls in a career, & spend entirely too much of the rest trading emails (my inbox can attest). Give them the bandwidth to run several programs at once, and the time to focus on scientific judgment, and I’m willing to bet they’ll make better calls. And at scale, better calls translate to more drugs, higher hit rates, broader patient impact, & likely, a disease or 10 or 100 cured. I couldn't be more excited to build toward that future @AnthropicAI. Finally, a heartfelt thank you to BioTwitter/X. The end of anonymity is bittersweet, but if anyone’s earned a proper introduction, you all have. It’s difficult to overstate how much this app has done for me: a decade of education, plenty of friends, many job opportunities, several mentors, a few investors, even a co-founder (fmr). It was the generosity of this community that inspired me to start posting 6 years ago, after just as many years lurking. I hope I’ve been able to pay it forward, in my own small way. But, to be clear, Prof Oak isn’t going anywhere. Same beat, same themes, and now, a front-row seat to drug development at the frontier. Stay tuned for dispatches from the trenches; and remember... Gotta Cure 'Em All
535
1,067
388
14,429
1,764,803
Siyu Chen 陈思羽 retweeted
Excited to share Bolero, out on bioRxiv. @arcinstitute posted a great overview. Here I want to add another perspective that motivated it for us (also a few more higher resolution images).
5
14
1
76
13,672
Siyu Chen 陈思羽 retweeted
Using AI to write for you - start to finish - is really just you robbing yourself of one of life’s greatest joys: to think through a problem and figure out what you think about it, and in the process becoming passionate enough to take on the new challenge of solving it.
17
36
5
290
14,069
Siyu Chen 陈思羽 retweeted
Integrating RNA-seq, metabolomics & proteomics, @GeraldineMaier mapped how exercise rewires metabolism across tissues: • Glycolysis • Fat oxidation • Ketone use • BCAA metabolism @WuTsaiAlliance cell.com/cell/abstract/S0092…
7
67
2
233
13,220
Yeah. Good scientific writing requires scientific thinking, reasoning, planning, and insights into both the broader field and specific research areas. None of these has a clear or easily verifiable standard. Even among humans, our ideas and interpretations are constantly challenged by different perspectives. After all, we should leave some of the fun of doing science to humans
Replying to @anshulkundaje
At the risk of being trite, verifying good writing requires human in the loop. We are far past RLHF. They are all about rapid loops on verifiable tasks of various time horizons. So the models do magic in the agentic coding space…. Improving writing will just be slower and at times at odds with the agentic mission.
2
72
Congrats Hani and Mehran and other team members on their new model and skill set
We assembled the Avengers to tackle the grammar of RBP-RNA interactions, especially how structure shapes who binds what: @MKarimzade, Lisa Fish, Heather Karner, @jjohlin, Ali Emadi, @drklly, @BoWang87, @hsnajafabadi, @LeonettiManuel
3
198
Wow, quite a similar research trajectory: synonymous mutations during PhD and tRNAs during postdoc. I think I know why I’m here, boss @genophoria
What is fascinating about academia as a career is that there are certain ideas that you have been working on since you remember. The numerical optimalities of the genetic code is what drew me to biology in the first place. I ended up working on codon usage as grad student and tRNAs as a postdoc, as the language layer on top of the genetic code. And now, 25 years later, I’m still amazed by how under-explored this field remains. So, new preprint with Vijay Ramani 🧵
1
230
Excited to share our new work on in vivo tRNA genetic dependencies! @genophoria @arcinstitute And Vijay @GladstoneInst Redundancy does not necessarily mean functional equivalence. Our findings reveal how tRNA requirements vary across tissues and environments, opening new questions about the relationship between genetic code decoding and biological context.
Hundreds of human tRNA genes form a redundant repertoire for decoding the genetic code. But does redundancy imply functional equivalence? A pan-cancer in vivo CRISPRi atlas from @genophoria and Vijay Ramani at @GladstoneInst reveals a structured landscape of tRNA dependencies.
1
4
394
Siyu Chen 陈思羽 retweeted
Excited to release AlphaGenome Atlas 🧬 We used AlphaGenome to predict the regulatory impact of all 9B possible SNVs in the human genome. We collaborated with amazing scientists to analyze and apply it, and developed a portal to browse the genome using this new lens 🔬 🌐 Portal: alphagenome.google/atlas 📖 Blog: goo.gle/4heuCvn 📄 Preprint: deepmind.google/blog/alphage… 🧵 1/6
11
86
11
348
36,898
Siyu Chen 陈思羽 retweeted
High-throughput omics are powerful, but they don’t substitute for thinking. Discovering new biology often starts with intellectual input: frame sharp questions first, then choose the methods that can answer them, not the one that’s most convenient or currently fashionable. 1/
Agree. But this is not a specific issue with omics. Comes down to a lack of serious consideration of expt. designs that can enable answering well defined questions. Vague problem definitions/formulations, arbitrary designs, scaling modalities just cuz u can. Huge waste.
1
14
75
11,699
Siyu Chen 陈思羽 retweeted
Agree. But this is not a specific issue with omics. Comes down to a lack of serious consideration of expt. designs that can enable answering well defined questions. Vague problem definitions/formulations, arbitrary designs, scaling modalities just cuz u can. Huge waste.
Unpopular opinion: the age of omics and high throughput data in Biology took us away from very good fundamental questions. The omics failed to have outsized returns as they were thought to and now we have a generation that’s trained to not think simple.
4
11
1
182
29,170
Congrats!
I’m excited to share our new study, “Spatial mapping of RNA turnover kinetics in the mouse brain,” published in Nature Neuroscience @NatureNeuro After six years of work, we now have a framework for studying RNA kinetics in living animals.
3
77