Ryan

Pinning

@seromics The fate of millions hangs in the balance

Fluge & Mella doing great work on daratumumab for ME/CFS - thank you to @ryan_czm for introducing me.

Did the daratumumab *just* destroy the myeloma - or did it *also* knock out a T-cell suppressing autoantibody? Daratumumab depletes long-lived plasma cells via CD38 & potentially-harmful immune-altering autoantibodies - not just the cancer! @casanova_lab

@JackHadfield14 Sir, when you measure across that many high dimensions with so few samples, any signal is likely to be noise.

One of the biggest reasons I’ve become theory-agnostic in Long COVID and ME/CFS is how easy it is to become convinced you’ve found the explanation. That is not because people are irrational. It is because the human brain has to simplify problems in order to handle them. When a disease is extremely complex, with many overlapping systems and huge amounts of literature, we naturally latch onto a theory that seems to connect a lot of the dots. I saw that in myself very early on with ROCK1. I came across papers suggesting it may be altered after COVID, and the more I read, the more compelling it seemed. ROCK1 appeared connected to mitochondrial dysfunction, vascular dysfunction, viral persistence, and a range of other features that looked highly relevant. It felt like I might have found something that could explain the disease. Interestingly, in our own data, ROCK1 appears reduced in blood. But even that is not straightforward. Lower circulating ROCK1 may in some contexts reflect cleavage, which could imply increased activity rather than decreased importance. That still needs further research. The same pattern repeated across other pathways. Angiotensin 1-7, ACE2, PINK1, TGFB, NAD+ biology, tryptophan and kynurenine metabolism, T-cell exhaustion, NK-cell exhaustion, and mast-cell biology. Each time, there were mechanisms that looked compelling. There were people who improved when targeting them. But there were also people who worsened from the exact same interventions. At a certain point, the pattern becomes difficult to ignore. Multiple pathways can each look capable of explaining the disease, which means likely none of them do on their own. Finally, we are beginning to see the emergence of systems that may be able to hold the full nuance and scale of biological data in context at once, something the human brain has not been able to achieve. Now the goal is clear. Build the highest-quality datasets to allow those systems to solve these diseases as quickly as possible 💪

Trump's approval rating just fell below 40 percent in our tracking for the first time. And his net approval rating is now -17.4, also a new low and down about 5 points over the past several weeks.

Iran War - Update : Negotiations are pure theater. Trump's 48-hour ultimatum was never credible. The Gulf states told him so directly as follow-through would have meant Iranian missiles on their own infrastructure. Iran offered nothing. Trump got a backroom readout from Witkoff, called it diplomacy, and bought himself two more weeks. Why two more weeks ? For troops to arrive ! ~7,000 US forces are moving toward the Persian Gulf right now. When they get there, expect ground operations. The target: Abu Musa, Greater Tomb, Lesser Tomb. Seize those, and the US gets a defensive foothold to escort convoys through. That's the optimistic version. Timeline: 2-4 weeks per Rubio. Reality: probably 4-6 week if all goes according to plan (it never does) Kharg Island stays off the table for now. You need 12-15k troops for that, and taking it likely triggers Iranian strikes on Gulf desalination plants. The Americans need to degrade Iran further before they go there. Trump's risk appetite has flipped. Oil heading to $100+ as a baseline, $200 not ruled out by Treasury. Rate cuts dead for the year. Midterms already a lost cause. So what do you do? You swing for legacy. Control the Strait. Leverage Iranian oil against China. Call it the "Strait of Trump." He literally said that out loud in Miami. Don't assume it goes smoothly. The Iranians have been sharper than anyone expected. That likely-hypersonic strike on the US base in Saudi Arabia on Friday wasn't nothing. The Strait is full of civilian boats and ringed by mountains, RPG-enthusiast heaven. Half the US Navy's amphibious ships are in poor condition with low readyness (41% mid last year). Iran absorbs this war and survives, but just about. IRGC tightens its grip on the state. Industrial base is gutted. Legitimacy collapses under the weight of repression, economic ruin, water crises, and incompetent new leadership. 3-5 year trajectory looks like either civil war along ethnic fault lines or a Pakistan-style military dictatorship. Neither is stable. Both are nuclear-adjacent problems down the road. Some but not all Gulf states come out stronger, paradoxically. Saudi Arabia is looking at +$50B/yr in oil revenue through Yanbu alone. GCC cohesion deepens. Defense spending surges. Long-term redundancy like extra pipelines to the Mediterranean/Red Sea, rail, water security, gets built in a hurry. China doesn't fire a single shot and wins. Stockpiles, state-directed subsidies, no midterm elections to worry about. Xi plays the long game, positions China as the "stable great power," and waits to be invited into the post-war multinational policing force for the Strait. Which will be a fascinating moment. Trump loses the House badly. Senate is a coin flip. What follows: investigations on every front (insider trading, anyone ?), a hobbled domestic agenda, and a president who then goes maximally unilateral on foreign policy — the one domain where he has nearly unchecked power. Ukraine gets cut loose in a side deal with Moscow. Europe erupts. NATO frays. Europe accelerates defense decoupling, builds toward its own nuclear umbrella, and quietly restarts the "variable geometry" strategy, waiting for the next US president, keep maximum optionality shopping between Washington and Beijing for whoever offers a better deal. End up commiting to neither. "Belle of the ball" strategy. This is not ending at a ceasefire. The positions are structurally incompatible. The economic damage is still being massively underpriced. And the escalation ladder still has many rungs left. Hang on to you helmets. This is the big one.

@tleilax___ Sir would elasticity dampen the ethanol factor slightly, if countries cut back on sugar use in a shortage more compared to staples like rice and wheat? Or would it be too minor of an impact

The US, the biggest gasoline market on Earth, just increased by 50% its ethanol blending limit. India is also considering to raise it from 20% to 30%. And you are short sugar because your Excel-spreadsheet analyst thinks the Brazilian sugarcane crop is half-decent ? ;) bloomberg.com/news/articles/…

@32Sfc46582 @scott_scientist When I say community I meant more researchers not patients, so yes.

J'ai compris mais dire que la communauté MECFS est idiote. Non juste désespérée, déprimée et on aimerait juste avoir des chercheurs à la hauteur de nos souffrances. J'ai l'impression que les chercheurs à la retraite et étudiants sur S4me.info sont plus compétents que les Ron Davis, Carmen Scheibenbogen, Prusty (ca c'est sûr) etc. Seuls Fluge et Mella et Ponting ont fait avancer la recherche. Le reste est nul.

I think I need to say this now for it to be believable later…. But beware of the first LC/ME trial that shows statistically significant improvement for an intervention. There are ways to mask symptoms and make someone look better, while causing them to deteriorate worse long term (years).

@32Sfc46582 @scott_scientist I'm not referring to you, sir. It is the idea that clinical trials out there that actually work aka F/M are not getting enough attention.

Pourquoi tu attaques la communauté MECFS ? Pourquoi être si condescendant ? La communauté MECFS est surtout désespérée. Avant de l'insulter, déjà respecte les plus malades. Tu as déjà été alité pendant des mois sans stimulation ? Tu sais ce que cela fait à un être humain ? Tout le monde n'a pas accès comme toi au Daratumumab, hein. Alors ne viens pas ici pour insulter les gens malades.

The world's electricity infrastructure, mapped.

The Brent Dated-to-Frontline (DFL) swap—is a critical instrument used to bridge the gap between the physical and futures markets. In simple terms, it reflects the price difference between Dated Brent (physical North Sea cargoes) and Frontline Brent (the first-month ICE Brent futures contract). In recent sessions, the DFL has surged into record backwardation, approaching USD 11/bbl and surpassing the previous peak seen in 2022. This sharp widening underscores an intensifying scramble for prompt physical cargoes, clearly highlighting the growing stress and dislocation in the market as buyers compete to secure immediate supply. Chart source: Bloomberg

Negative prices essentially is a signal that says the market at this time is irrelevant because generation that has contracts is such a massive amount in the system. It says nothing about who the holder of those contracts are and if they are actually saving money on average. They could be the public agreeing to pay a wind farm £150 a MWh for example on a legacy FiT.

Thailand, Brazil and India as very exposed to the loss of Gulf nitrogen fertilizers. Here is an assessment of the potential yield losses. This is using simple yield response models based on FAOSTATs data.

JEPXスポット市場2026年3月28日(土)受渡分取引価格☞システムプライス平均値📊10.57円／kWh、最高値📈18.56円／kWh、最低値📉0.01円／kWh、東京BL🏙12.12円／kWh、関西BL⛩9.31円／kWh、取引量⚡778GWh＜天気予報＞札幌🌧10℃／4℃、東京🌧20℃／11℃、大阪☀22℃／9℃、福岡☀20℃／10℃

Escalated with China and found out that rare-earths are leverage. Escalated with Iran and found out that Hormuz is leverage. About to escalate with Europe and find out that net direct foreign investment, military bases, IP of the tech sector, advanced tooling and pharma are leverage too.

@SalvMattera It did do its job, which is boost my PBMC NK cells from 50 to 420 in a single week after a 2mg dose.

I really don't know anything about it, but I'm always skeptical of magic beans that claim to treat everything

The discussion about this drug in the Long COVID space has always felt very strange to me.

To all those saying "This is good..." When Americans want to visit Japan, learn the BOJ's moves/dynamics, study Japanese direct drive turntables, or bamboo baskets (there are literally more collectors of Japanese bamboo baskets in the US than in Japan), they might read my

@_apedev @Bert_with_ME @TheGregYang you are a smart guy (a genius) but you have fallen for the Lyme grift. Don’t repeat their mistakes.

@Bert_with_ME Lyme disease is a whole can of worms. Most of the quacks who practice in it just prey on vulnerable people and rely on the fact that with Lyme it's difficult and expensive to prove a negative.

In the early days of my illness, I went to a prominent LLMD who clinically diagnosed me with Lyme & Babesia simply based on symptoms and history without running a single test. He prescribed me extended courses of Doxycycline + Malarone. My sense was that 9/10 patients who walk through his door get the same or similar diagnosis and treatment. The fact that he did not feel the need to screen for autoimmune disease, thyroid stuff, reactivated viruses, etc. prior to making that diagnosis was a glaring red flag. Months later, I ended up shelling out $2K for the full tick-borne panel from IGenex, and it was stone cold negative across the board. I have a lot of empathy for Lyme patients, but the scene is more sketchy/grifty than even ME/LC from my brief experience.

Brazil and India are the world two largest producers of sugar. Sugarcane is an extremely voracious crop in term of nitrogen per hectare. Russia just banned nitrogen fertilizer exports and Brazil + India were the biggest buyers of Russian fertilizers. youtu.be/h50eAv9tjvA?si…

@cloneofsimo From a Lin Alg POV, I guess LLMs wander around in the spanning set of knowledge. They can’t add more dimensions to the set.

Its very possible that LLM trained on newtonian physics may never come up with relativity to explain cosmic scale gravity. In that case Einstein would have to intervene and solve it instead. But would he have had come up with it, assuming he offloaded all the physics problem solving to LLMs? I think this is serious problem. Undoubtedly many GOATs are only GOATs because they built all the intuition from problem solving themselves. Grothendieck famously reinvented measure theory from scratch when he was teenager. If people offload their RL envs they couldve used, to LLMs, we will never get the next Einstein

@EmanuelDerman There are some masters of this on Twitter/X, especially in commodities/finance. I'm just here to learn from them :D

@ryan_czm Tough one

The most interesting thing in the world is trying to see if you can understand what drives other people, putting yourself in their place and mind