Gregory R. Johnson retweeted
The @newlimit team will be hosting a special Progress Update tomorrow at 5:30pm Pacific.
Goal is to share more about early work to try and extend human healthspan using epigenetic reprogramming.
Tune in to the livestream here:
youtube.com/watch?v=kc6gcJKb…
Gregory R. Johnson retweeted
For those curious about our quest @newlimit to reverse aging. We are presenting our work over the last 6 months and our plans over the coming years in a livestream at 5:30pm pst tomorrow 4/13. livestream.newlimit.com/
Gregory R. Johnson retweeted
Manual inspection of data has probably the highest value-to-prestige ratio of any activity in machine learning.
Gregory R. Johnson retweeted
The first main open-source product from our group is out on arXiv (arxiv.org/abs/2209.02498). We developed a generic image-to-image transformation package that be directly applied on 2D/3D/4D data for semantic segmentation, instance segmentation, image restoration, ... 1/5
Gregory R. Johnson retweeted
Introducing scPerturb, a collection of 44 single cell perturbation datasets with unified annotations available at scperturb.org (1/5)
Gregory R. Johnson retweeted
The partial reprogramming preprint from @JacobKimmel + colleagues last year has been published in @CellSystemsCP.
Revisiting it, I'm fascinated by 5 observations.
Gregory R. Johnson retweeted
our work on diverse partial reprogramming strategies is out today @CellSystemsCP!
cell.com/cell-systems/fullte…
Gregory R. Johnson retweeted
excited to share that I'm helping to build @newlimit as the Head of Research
we're developing a new class of reprogramming medicines for age-related diseases & recruiting broadly across diverse fields of science for our research division:
newlimit.com
Gregory R. Johnson retweeted
Happy birthday to Buffy Sainte-Marie, born on this day in 1941 on the Piapot Cree First Nation Reserve in Saskatchewan, Canada. Here she is performing "Cripple Creek" on a mouth bow on the children’s television show Sesame Street in the late 1970s.
Gregory R. Johnson retweeted
When two wrongs make a right...
(Antibodies with different biases can be used to produce an unbiased measurement through prediction. Don’t postprocess; train.)
This Post is from an account that no longer exists. Learn more
Gregory R. Johnson retweeted
What I fail to understand is why millions of (public) money are spent on (medical) research without (properly) employing statisticians to ensure the quality. It is like letting pilots build planes after having coffee with an engineer. No one would ever fund let alone trust that.
Replying to @GlassockJ
Unfortunately, methodologic peer review in medical journals is so poor that a large portion of published medical research should simply be ignored. It's hard to know what proportion of findings are harmful, but I'm confident that a large proportion is uninterpretable.
Gregory R. Johnson retweeted
modern biologists use dimensionality reduction all the time. variational autoencoders find a lower dimensional spaces with some nice features: continuity, interpolation, & uncertainty.
I wrote an introduction for curious CompBio folks:
jacobkimmel.github.io/variat…
Gregory R. Johnson retweeted
Join us and @Blizzard_Ent this Thursday at 6:00pm GMT for an exciting #StarCraft demonstration, hosted by @Artosis and @RotterdaM08!
Livestream on YouTube: youtube.com/c/deepmind
Read more about #StarCraft2 as an environment for AI research: deepmind.com/blog/deepmind-a…
Gregory R. Johnson retweeted
.@Allen_Institute scientists are using #GPUs + #AI to predict the look of stem cells to better understand cancer. nvda.ws/2nTonjJ
















