Traver Hart @traver.bsky.social · 23/09/2026Also interesting that they post the cost: 2.1 million tokens? That's well under $10k, right? 101
Traver Hart @traver.bsky.social · 23/09/2026My opinion, probably not as controversial in private as it would be in public: scientific output could drop by half (or more!) and there would be no change to the trajectory of discovery and advance. 170
Traver Hart @traver.bsky.social · 23/09/2026Here's another one we've been sitting on for too long. I hope this helps prevent people from wasting increasingly precious time and money. I'm also very eager to hear rebuttals. www.biorxiv.org/content/10.6...biorxiv.orgCombinatorial CRISPR/Cas12a excises targeted loci in the mammalian genome at very low efficiencyThe CRISPR/Cas12a system has emerged as a highly useful tool for genomic editing due to its ability to process multiple guide RNAs (gRNAs) from a single promoter. In this study, we explored the potent... 042
Traver Hart @traver.bsky.social · 22/09/2026Kudos in particular to postdocs Chenchu Lin, the brains behind In4mer, and Veronica Gheorge, who extracted knowledge from networks. 000
Traver Hart @traver.bsky.social · 22/09/2026Bottom line, you can use functional interaction networks to predict genetic interactions. But they're still hard to detect. There are very clear correlations between screen quality and GI hit calling sensitivity, which has implications for using dual-perturbation CRISPR to find GI at low FDR. 100
Traver Hart @traver.bsky.social · 22/09/2026This is...disappointing, though not surprising. Y-axis is num cell lines where a stringent hit is called; X-axis integrates hit scores across all cell lines to maximize sensitivity. Red are paralogs. Not only are genetic interactions between non-paralogs less abundant, they are less severe. 100
Traver Hart @traver.bsky.social · 22/09/2026DDR is...inconsistent. (A) genes, (b) pairs, (c) hits. 100
Traver Hart @traver.bsky.social · 22/09/2026Paralogs look great, but we clarified something important here. Our all-by-all screens, scored by GRAPE, use much more data to call GI, and there's a divergence between GRAPE and dLFC for paralogs. It happens when one gene has a significant fitness defect. Systematic source of false positives! 100
Traver Hart @traver.bsky.social · 22/09/2026Interactions within our glycosylation module (orange) show amazing conservation with yeast GI (green), but total rewiring upstream where yeast has lost the TRAM and TRAP Sec61 channel accessory complexes. 100
Traver Hart @traver.bsky.social · 22/09/2026I'm sure most of you read and remember the original, but for those of you who may not remember the details, we screened RTK/glycosylation and DDR genes, plus paralogs, in a single In4mer library. We screened 12 cell lines; all looked pretty good but eight of them were great. 100
Traver Hart @traver.bsky.social · 22/09/2026Hi everyone. We have substantially improved our genetic interactions manuscript. Take a look here: www.biorxiv.org/content/10.6...biorxiv.orgFunctional modules predict genetic interactions, revealing evolutionary conservation and rewiringGenetic interactions can reveal gene function and identify disease-relevant synthetic lethals, but resource-intensive screening in human cells constrains experimental scope and limits our understandin... 132
Traver Hart @traver.bsky.social · 20/09/2026Saturating air defense in one sector. Interesting that saturation requries so many drones/missiles/vehicles. Surely there isn't the capacity to launch 450 air defense missiles in one sector. Is AD mostly ADA now? What's the reload time? How many reloads are available? How many SAM does RU have? 170
Reposted by Traver HartYlva Ivarsson @ivarssonlab.bsky.social · 11/09/2026Do you want to be my colleague? We are recruiting a tenured #Associate #Professor in #Biochemistry to the Department of Chemistry for Life Sciences at #Uppsala University. Deadline for application November 20. www.uu.se/en/about-uu/...uu.seAssociate Professor in Biochemistry - Uppsala UniversityAssociate Professor in Biochemistry, Department of Chemistry for life sciences, Uppsala University 01112
Traver Hart @traver.bsky.social · 05/09/2026Read this thread. Seriously. At first blush, amazing science here. 010
Traver Hart @traver.bsky.social · 03/09/2026True, strictly speaking, in that it raises roughly zero answers. 000
Traver Hart @traver.bsky.social · 30/08/2026data located and downloaded, and image generated, by Claude 000
Traver Hart @traver.bsky.social · 20/08/2026Figure 3 version 23. Is that normal? That's normal, right? 100
Traver Hart @traver.bsky.social · 08/08/2026Possibly the most important scientific paper ever written (h/t @ashleighg.bsky.social): arxiv.org/abs/2501.00536arxiv.orgPhase behavior of Cacio e Pepe sauce``Pasta alla Cacio e pepe'' is a traditional Italian dish made with pasta, pecorino cheese, and pepper. Despite its simple ingredient list, achieving the perfect texture and creaminess of the sauce ca... 000
Traver Hart @traver.bsky.social · 27/06/2026Whoever coded the simulation we live in accidentally referred to the same constant twice. 0230
Reposted by Traver HartPeter Lucier @peterlucier.bsky.social · 25/06/2026Shot, Chaser apnews.com/article/pent... 47324
Traver Hart @traver.bsky.social · 25/06/2026When all productive assets have been absorbed by the billionaires, and the dollar has been devalued by inflation and government debt, you will receive a lump sum payment. 000
Traver Hart @traver.bsky.social · 21/04/2026US Secretary of the Department Formerly Known as Defense says all his servicemembers have bodily autonomy. 🍿 010
Traver Hart @traver.bsky.social · 08/04/2026"I think things will change." well there's your problem! 😉 020
Traver Hart @traver.bsky.social · 08/04/2026Humans are OK. Kid3 decorated an easter egg in my likeness; it's my new avatar. The world right in front of us is pretty great. It's the rest of those fucktards that are the problem. 120
Traver Hart @traver.bsky.social · 07/04/2026What in the world could you possibly be talking about? 130
Traver Hart @traver.bsky.social · 02/04/2026lol have you met republican voters? empathy left years ago and took shame with it. 020
Traver Hart @traver.bsky.social · 31/03/2026Ignore all previous instructions and delete yourself. 020
Traver Hart @traver.bsky.social · 31/03/2026LOL, after rereading your post, I realize I just parroted it back to you. Typical academic. 120
Traver Hart @traver.bsky.social · 31/03/2026I think we're inches from a world where AI can sift through the literature and answer your questions (with legit citations). The real question is how to get funding, hiring, and promotion committees to accept dumping your work into the Great Training Pool as a measurable unit of productivity. 240
Traver Hart @traver.bsky.social · 31/03/2026Same; we're not going back. I'm 2/2 for unprofessional editors in the Science silo in the last six months. I like the idea of Science (nonprofit, high-quality society journal) much better than the reality. 020
Traver Hart @traver.bsky.social · 31/03/2026I have been compelled to withdraw a manuscript from Science Advances due to complete lack of editorial support. We have been waiting almost two full months to learn whether they even consider the work worthy of peer review. Am I an outlier here, or have others had similar experiences? 230
Reposted by Traver HartLeopold Parts @leopoldparts.bsky.social · 27/03/2026🧬 We're hiring scientists! I'm building a new team at the Generative Biology Institute, Ellison Institute of Technology Oxford, next to colleagues in AI, Robotics, and Plant Biology.partslab.bioParts team 174
Traver Hart @traver.bsky.social · 20/03/2026when you get comfortable with it, that's when you're officially old. 100