Yuge Ji (@_yji_) profile picture

Yuge Ji

@_yji_

ML in compbio, scanlating~

 Cambridge, MA
Joined February 2016
  • Tweets73
  • Following241
  • Followers218
Do you run functional assays? Wish you could get more results without having to scale? If you’re not using Prophet, you’re leaving potential on the table. (Warning: pitch not tweetorial)
1
13
3
33
11,042
Sometimes I think about how mostly male modern media sold a version of female beauty and then women stopped wanting to have kids in part because it conflicts with those beauty standards and, well, if it isn’t the consequence of your own actions
92
TIL if you fly on your birthday, your #United flight attendant will surprise wish you a happy birthday and give you a snack
127
Gave a presentation at the MassBio summit yesterday and someone asks me how to interpret the "piles of dust" on my slides. I now have a new favorite term for UMAPs
3
273
I went to #BIOInternational this year and wished I'd gone 10 years ago. This is my attempt at convincing any early career researcher(or otherwise) to go for this unsupervised data collection experience. open.substack.com/pub/yugeji… #BIO2025
Image from Yuge Ji (@_yji_)
3
271
Meet the judges & mentors for the Tahoe Deep Dive Hackathon & thanks @awscloud for offering the prize: $25K AWS credits to winner, 10K & 5K to 2nd and 3rd. Deadline to apply: Apr 4th, details in 🧵. A short DM with @ClementDelangue; and now an entire community coming together!
Image from Nima Alidoust (@nalidoust)
4
11
4
42
19,744
Yuge Ji retweeted
Peter Koo (@pkoo562) profile picture
Peter Koo@pkoo5625 Feb 2025
[SAVE THE DATE] MLCB 2025 is happening Sep 10-11 at the NY Genome Center—NYC! Attend the premier conference at the intersection of ML & Bio, share your research and make lasting connections! Submission deadline: Jun 1 Details: mlcb.github.io Spread the word—please RT!
1
41
1
109
12,880
Hot take: if you work in AI for medicine you should be able to pass Step 1
1
1
176
And no, I cannot
154
Grateful to see industry putting out thoughtful benchmarking work like this biorxiv.org/content/10.1101/… but sad to imagine what they must think about academia and bio ML papers
Simple controls exceed best deep learning algorithms and reveal foundation model effectiveness for...

Simple controls exceed best deep learning algorithms and reveal foundation model effectiveness for...

Modeling genetic perturbations and their effect on the transcriptome is a key area of pharmaceutical research. Due to the complexity of the transcriptome, there has been much excitement and develop...

biorxiv.org
1
1
4
257
Pfizer (the authors here) is the end user, they shouldn't be the ones having to call out papers from @NatureBiotech What is the point of academic research if the hurdle of "trusting it enough to try it" gets so big no one uses academic methods at all?
1
1
2
343
I know the problem lies with how publishing incentives work but I wish there were a way to repair trust, even despite the current publishing system
1
140
1/7 Planning to build a single-cell atlas? Or wondering how atlases can be useful to your research? Read our guide on single-cell atlases nature.com/articles/s41592-0… published today in Nature Methods, by @SikkemaLisa, @KHrovatin, @MDLuecken, @fabian_theis and others.
Image from Lisa Sikkema (@SikkemaLisa)
1
16
3
54
5,668
Guys let’s move to Japan
4
4
413
And probably add a few years to your lifespan on top of everything else (come through longevity twitter)
1
233
(If you have kids, they’ll be multilingual!)
186
Live there for a few years, return fashionable AND polite
179
All you have to do is learn a little bit of Japanese
159
11/14 Looking ahead, we believe progress in this field will specifically require two key elements: - Higher-quality data spanning a wider range of cellular states and perturbations - Specialized models designed to fully leverage large-scale datasets for perturbation prediction
1
1
2
574
First (public) blog post - on single-cell, bubbles, and discovery yugeji.substack.com/p/half-a…
Image from Yuge Ji (@_yji_)
1
4
223
My silver lining is that maybe this means more funding goes to high quality data generating institutions and technologies, the only reason we have AI in the first place
4
412
Load more