Spurious correlation inflates performance in single-cell perturbation prediction
www.biorxiv.org/content/10.6...
But I suspect that we will see the idea that clear writing and concise articulation of ideas is some kind of quality differentiator die. And there are those who will
say it was always a poor quality proxy, favoring only good writers not good scientists.
"Hypothesis generation: pretty good...data analysis: a super power...executing experiments: not so much"
AI can help at every stage of the research cycle. Jonah Cool from @anthropic.com discuss the benefits of integrating frontier models and potential concerns youtu.be/sz1vQyBzSgw?...
"a paper today is a Frankensteined-together patchwork of supplementary files that run longer than the manuscript, datasets in repositories...code posted to GitHub without documentation. The paper has been replaced....by a mess" www.thetransmitter.org/from-bench-t...
Introducing openRxiv Labs! 🔬 Growing from the strong foundation of bioRxiv & medRxiv, Labs is an experimental space for working with collaborators to push the boundaries of open scientific communication.
👉 Check it out and learn more: openrxiv.org/openrxiv-lab...
#OpenScience #Preprints #Scicomm
Excited to share our RegVelo paper in Cell
www.cell.com/cell/fulltex...
We unify RNA velocity + GRNs into one model → better OOD prediction of perturbations (e.g. gene KOs), with examples incl. neural crest KO predictions 🔬
Big thanks to W Wang, Z Hu & T Sauka-Spengler 🙏
One of the good things about the AI co-scientist brouhaha is that it makes us discuss what science *is*, and how to measure its value, impact, quality. — Questions that have been there all along, but many preferred to gloss over.