Indeed, #methodsmatter nature.com/articles/s41592-0…

Interesting blog post by @wtgowers on (training to do) mathematics and AI. His conclusions hold equally well for computational biology. gowers.wordpress.com/2026/05…

⚠️All of the below images were FABRICATED by ChatGPT Images 2.0, each with a single prompt❗️ ⚠️

.@ENCODE_NIH Phase III (2020) didn’t deliver any splashy slogan like the 2012 “80% functional genome” bullshit. Instead, it quietly retreated. Now we’re told functional annotation is “still in its infancy.” No wonder it took me 6 years to notice that ENCODE 3 even existed. nature.com/collections/dggcc…

The ρPCA method is implemented in an easy to use Python package github.com/pachterlab/rhopca… Quick to install. Fast. We now use it routinely whenever engaging in exploratory data analysis. 8/

Actually, your analysis might fail regardless of which you picked. But one thing is for sure: your analysis will be very different depending on the choice. sciencedirect.com/science/ar…

This latest article shows how AI researchers themselves are blind to their own hype. Their overconfidence demonstrates the opposite of what good scientists should be doing. I'm deeply skeptical that AI can do any useful research right now, and this article doesn't change my mind

Correct interpretation of p > 0.05: The current sample size did not yield sufficient evidence to reject, at the incredibly arbitrary 0.05 level, the supposition that the treatment does nothing, assuming data model M (describe M ...) which forgot to include covariates #Statistics

Peer review was supposed to be science’s quality filter, but somewhere along the way it started acting more like a bouncer who only lets in the regulars. It’s slow, it tends to favor established labs and familiar names, and it gets uncomfortable around anything too unconventional. Papers loaded with mountains of data tend to cruise through, while bold ideas that actually challenge the consensus get stuck in limbo or turned away at the door. The irony is that where a paper gets published almost never determines its real worth. What actually matters is what the scientific community does with it afterward, whether people cite it, argue with it, build on it, or use it to blow up a long-held assumption. That’s where the value lives, not in the journal’s logo. A major survey a few years back found that roughly 70% of researchers think the current system is fundamentally broken, and it’s not hard to see why. Publicly funded research hides behind paywalls, editors chase whatever topic is hot that month, and the whole incentive structure pushes toward safe bets over genuinely risky and potentially important work. Science has always been complicated and deeply human and full of ego and inertia, but the conversation is shifting.

kallisto sped up RNA-seq quantification by 50x. Now another 50x speedup... quantify hundreds of millions of reads in a few seconds. This seems too good to be true. But it's true! Incredible accomplishment by @pmelsted.

I'm happy to share this free link to our new paper on the perils of trying to find microbes in human cancers, which appeared today in @NatureCancer. Co-authors include Nobel laureate @barjammar and ancient DNA expert Eske Willerslev: rdcu.be/e4IaU

I used Claude Opus 4.5/4.6 (and a bit of Codex GPT-5.3) to port edgeR to Python. See edgePython github.com/pachterlab/edgePy… This allowed me to develop a single-cell DE method that extends NEBULA with edgeR Empirical Bayes. All in one week. Details in doi.org/10.64898/2026.02.16.…

To perform machine review we did not just feed papers into an LLM. We designed a claim extraction system, an approach to group related claims into results, and a way to compare results directly. Examine the comparisons for the whole @eLife corpus here: openevalproject.com/ 2/

Spatial transcriptomics is powerful—but fragile. Our new @NatureGenet paper shows that cell-segmentation errors are widespread and can obscure much of what we think we’re learning. nature.com/articles/s41588-0…

The fact that some genes are only "expressed" after death does not teach us anything about function. It only shows that that this method of studying gene expression is completely and utterly useless. the-scientist.com/how-brain-…

Comprehensive benchmark of differential transcript usage analysis for bulk and single-cell RNA sequencing academic.oup.com/nargab/arti… 🧬🖥️🧪

New blog post: In Google's new paper, "An AI system to help scientists write expert-level empirical software" an AI supposedly "invents" new methods in bioinformatics and infectious disease modeling & forecasting. doi.org/10.59350/ymjnw-nj555 🧬🖥️🧪

In a new work with @Josephmrich and Conrad Oakes we tackle the problem of how to best organize alluvial plots. We formalize two optimization problems and develop a solution for them based on the neighbornet algorithm, implemented in the program wompwomp: github.com/pachterlab/wompwo…

The spacexr package for cell type identification in spatial transcriptomics is now on Bioconductor! Includes a fast, efficient implementation of RCTD Future updates: CSIDE reference-free version Link in the reply 👇

GPT-4o image generation. "Make the bound RNA band under F3 look more intense and well defined" Original vs GPT-4o fraud. (Original Image Fig 4E from 10.1016/j.molcel.2025.03.006)