Bioecomatics

#STDANDARIS2DQUANTPROTOCOLS #MCQiQCM #aiGEN2SPEC2ULTRASPEC #FIXEDPROPORTION #PARITY

Presented at #ACC26: Among patients with atherosclerotic cardiovascular disease, targeting an LDL cholesterol level below 55 mg per deciliter led to a lower 3-year risk of cardiovascular events than targeting a level below 70 mg per deciliter. Full Ez-PAVE trial results: nejm.org/doi/full/10.10… Editorial: Paving the Road toward Targeted Lipid Lowering nejm.org/doi/full/10.10… @ACCinTouch

"Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?" Self-distillation can make LLMs look smarter by producing shorter, more confident reasoning traces, but in math it often takes out the model's uncertainty and self-correction signals. This can potentially badly hurt out-of-domain reasoning, so this paper suggests that better reasoning is not just about compression, but preserving useful "doubt".

This paper took us 13 years and is one of the longest papers ever in Cell. Check it out & judge for yourself cell.com/cell/fulltext/…

GLP-1 and the cardiovascular system "This Review summarizes the effects of GLP-1 and GLP-1RAs in the CV system..." jci.org/articles/view/…

US President Donald Trump has named 13 people to his panel of science advisers — and all but one is a leading technology executive. go.nature.com/4uWojBD

Mitochondria emerging as leading immune organelles

I've recently stopped accepting manuscript review requests. With today's top AI frontier models now able to do a better job than me in the majority of cases, my contribution is usually minimal. H/t @DeryaTR_

AI will increase productivity, reallocate labor from routine tasks to technical tasks and cause large (small) firms to reduce (increase) workforce, from Salomé Baslandze, Zachary Edwards, John Graham, Ty McClure, Brent H. Meyer, Michael Sparks, Sonya R. Waddell, and Daniel Weitz nber.org/papers/w34984

1 in 5 people carry a genetic heart attack risk that no diet, exercise, or statin fixes. It's called lipoprotein(a). Only 0.3% of Americans have ever been tested. The paradox: statins, the most prescribed cardiac drug on Earth, raise it by 10-20%. What actually works today? 1️⃣ PCSK9 inhibitors: best available. 25-30% Lp(a) reduction (FOURIER, n=25,096). Proven CV benefit. 2️⃣ Inclisiran: siRNA targeting PCSK9 mRNA. ~20% reduction, only two shots per year. No outcomes data yet. 3️⃣ Ezetimibe adds ~7%. Niacin drops the number by 23% but two large trials found zero fewer heart attacks. 4️⃣ Apheresis hits 70-85% per session but it's a 3-hour blood filtering procedure every two weeks. Five drugs in late-stage trials are hitting 80-94%. Lepodisiran leads at 94% (Phase 3, Eli Lilly). One, muvalaplin, is a daily pill. Behind them: CRISPR gene editing. A single dose of CTX320 cut Lp(a) by 94% in primates. Lasted 7+ months. Phase 1 human data expected this year. The gap: 25% now. 94% one-time cure coming. One blood draw to find out if you need it.

In three RCTs, #Tirzepatide was associated with clinically meaningful weight loss among adults with overweight or obesity, even when concomitant weight-inducing medications were initiated during treatment. ja.ma/4lWXYzt

🚨 BREAKING: Someone built an AI that replaces 6 months of literature review with a single workspace. It's called Moara. You dump your research papers in. The AI organizes, analyzes, and synthesizes everything across hundreds of millions of academic sources. You come back and there's a structured literature review waiting. Not a chatbot answer. Not a vague summary. The actual research synthesis. A full research workspace that thinks like a scientist. Works for medicine, engineering, social sciences, law, business, humanities — every discipline. Here's what it does on its own: → Searches hundreds of millions of papers across PubMed, arXiv, Google Scholar, Semantic Scholar, ClinicalTrials .gov, Cochrane, and more in one shot → Screens and tags results automatically, pulling out methodologies, findings, and contributions from every paper → Lets you chat with your entire library at once, find themes, spot gaps, and compare findings across dozens of papers simultaneously → Integrates with Zotero, LibKey, RIS, and BibTeX so your existing research workflow doesn't break → Traces every AI insight back to its original source so nothing is hallucinated and everything is verifiable → Supports full team collaboration so entire research groups can screen and synthesize together Try it here: moara.io/?utm_source=rb

Really exciting to see KV-cache compression getting attention. A similar bottleneck shows up beyond LLMs: for world models and autoregressive long-video generation, KV cache can quickly dominate memory and limit long-horizon consistency. Our recent work, Quant VideoGen, explores training-free 2-bit KV-cache quantization for video diffusion models, achieving up to 7.0× KV memory reduction with <4% latency overhead. Link: arxiv.org/abs/2602.02958

AI inevitably leads to this, in every country. IMO, this is more scary than nuclear. One hacker ... Sadly, I don't see a way to avoid it.

Substantial advances in radiation therapy have significantly reduced cardiac dose. But vigilance remains key—screening for CV risk & disease is critical. #CVD is detectable, treatable, & preventable. Explore more: jacc.org/journal/cardio… @DrBonnieKy #ACC26 #JACCCardioOnc

During a keynote session at APA 2025, Dr. Margaret Morris unpacks the risks of “sycophantic” AI: tools that validate everything, even when challenge and reflection are what’s needed most.

Still creating your reports manually in Word or PowerPoint? This is one of the biggest productivity bottlenecks in data analysis because every change in your data requires manual updates to results, tables, and visualizations. This process is not only time consuming but also increases the risk of errors. ❌ Results can quickly become outdated when the data changes. ❌ Copy and paste workflows increase the chance of mistakes and inconsistencies. Quarto solves this problem by allowing you to create fully reproducible and automated reports directly from R. The visualization below shows the result of a typical Quarto workflow. You combine text, R code, and output in a single file, and Quarto automatically generates the final report. This ensures that your results, tables, and visualizations are always up to date and fully synchronized with your data. I’ve just published a new module in the Statistics Globe Hub on how to build reports with Quarto in R. It includes a full video walkthrough, practical examples, and exercises that take you step by step through creating reproducible reports. Not part of the Statistics Globe Hub yet? The Hub is a continuous learning program with new modules released every week on topics such as statistics, data science, AI, R, and Python. Join now to access the Quarto module and all other modules released this month. More information about the Statistics Globe Hub: statisticsglobe.com/hub #rstats #datascience #statistics #quarto #reporting #reproducibility #statisticsglobehub

Genetic modifiers of APOE-ε4-associated cognitive decline nature.com/articles/s4146…

Functional connectivity-based attractor dynamics of the human brain in rest, task, and disease elifesciences.org/articles/98725…

New online! Cardiovascular stent technologies for coronary and valvular heart disease: the potential of 3D printing for stent fabrication dlvr.it/TRkhwJ

Oil’s global ins and outs @ShaleTier7 explains the energy situation for tech people: a16z.news/p/the-energy-s…