Save NASA GeneLab 🧬 ! If you care about space 🚀, or medical applications here on Earth 🧠, or both, help preserve GeneLab’s impact on biomedical discovery and open science, and its necessity for space exploration. www.change.org/p/save-nasa-...
DrugDomain2.0: comprehensive database of protein domains-ligands/drugs interactions across the whole Protein Data Bank www.biorxiv.org/content/10.1101/2025.07....
I love papers comparing pipelines! They’re under-appreciated, and we could use more of them, in my opinion.
I think most of us have a list of stock responses to review points that everyone knows are polite ways of saying, “editor, you & I both know you should just drop this question from the exam.” But it’s always awkward when you have to do it for an entire review…
Grammatical spite as a longevity promoter seems more promising than most of what people are peddling, tbh
Sometimes I suspected one of my employers intentionally somehow circumvented Outlook rules. I could never make enough rules to catch all the spam newsletters sent by random, comically vague-sounding university centers/cores/lists.
I love the USCPSC.
I’ve had to become a “serial specialist,” in one field, then the next, etc, trying to use past expertise along the way, rarely given space to do so to my satisfaction. Finding settings allowing their application to problems I think their intersection may help w is the biggest challenge of my career.
Agreed. But hopefully we work towards a society where higher and higher percentages of doctors *can’t* be replaced, which requires that they be given the support/resources/systems needed to be irreplaceable.
"Any doctor who could be replaced by a computer, should be replaced by a computer" ~ Jonathan Chen's talk at #psb25
Questionable practices in machine learning arxiv.org/pdf/2407.12220 #machinelearning #artificialintelligence #datascience
Photo of a man holding the book Hidden Potential
I'm reading Hidden Potential by psychologist Adam Grant. I put together a list of my favorite insights from the book to help you unlock your hidden potential.
Check out the thread below, and if you find it helpful please share and follow!
🧵
#booksky #psychology #self-improvement
Naughty Chair by Connie Jessop A cartoon child sits sadly in a chair next to a dog gently napping and a red ball behind him.
In fact, we use this kind of operant conditioning in how we raise kids: we link positive outcomes to prosocial behaviors, negative outcomes to antisocial behaviors.
But this leaves us vulnerable to thinking that is hyper-detection of causality, even in random, unlinked events.
why is that financial company spelled Schwab and not əb
That’s a big question! You could fill a whole library with research in that area. One researcher whose work comes to mind and that I would recommend is Daniel Foti.
Sure!
I'm back on Youtube after nearly a decade! I'd be super grateful if you watched my latest video and leave a comment letting me know what you think! It's on the imposter syndrome, and how the anxiety from imposter syndrome can actually fuel genuine confidence. youtu.be/qwlmlhFWv3w
variable names *are* code comments
If anyone already has access to this paper, please send it my way!
I can’t wait for this to finish processing in bioRxiv and become available. Mapping spatial omics data from separate samples onto a common tissue coordinate system so they can be analyzed jointly (e.g., for condition/group comparison) is 1 of the most pressing needs in biology right now in my view.
Still reading through it but seems to be a neat ex. of using spatial tx’omics + IF, scRNA-seq/Tangram, & RNA velocity/spatial pseudotime to not only map landscape but mechanisms by probing cell-cell ixs. Substantively interesting, but may also be of interest to folks looking to combine these methods
Many would consider “I was trying to install CUDA drivers” a horror story in 7 words
I never did master the art of convincing people that UMAP white space does not mean what it tempts you to think it means. Has anyone succeeded at communicating this?
I just wrote a new post on my substack! It's about love, fear, control, and Christmas trees. Please take a look and let me know what you think :-) writing.danielwendler.com/p/how-to-lov...
It’s freeing!
I highly recommend copulas for multi-variate models with varying distribution types: stat.uiowa.edu/sites/stat.u...
Simulation is really the only avenue for most of the models I use! Also good for violations of distributional assumptions.
Cell types as scientist-imposed naming schemes that, rather than being some ground-level truth, summarize useful information/group similar entities, is the part of #biology that most reminds me of the challenges of “fuzzy” aspects of #psychology research (e.g., naming factors of psychopathology).
I think many people may benefit from the availability of a multi-day workshop in transitioning from being an advanced end user to methods dev. Many fields are short on people (particularly substantive experts) developing methods, in part because there’s little guidance on making that jump.