
MD/PhD student at @ChariteBerlin & @MSKCancerCenter in @CalebLareau lab
New York, New York
Joined September 2019
- Tweets39
- Following263
- Followers171
- Likes84
Laura Kida retweeted
RFD4-Proteina does not do well for the application you care about, but you are not an ML expert?
Don't worry, @RisingSayak and the @huggingface team have you covered: easy LoRA finetuning directly in the diffusers library! Just make a dataset of interest with atomworks and go!
You can now LoRA fine-tune RFD4-Proteina from @NVIDIAAI with Diffusers for protein design 🤯
The example trains rank-32 adapters on the denoiser with flow-matching losses (on backbone coordinates + local latents instead of pixels).
Most of the training code is very similar to how we do for standard image and video generation models.
Details:
github.com/huggingface/diffu…
Laura Kida retweeted
Come check out our first preprint on the most interesting viruses that no one has heard of, anelloviruses! With @JacobJacobog02 and @ThahminaAli we delve into the determinants of anellovirus infection and persistence in humans
Our bodies are colonized by trillions of viruses that we live with in homeostasis. The most diverse and abundant of these viral families are anelloviruses. A new paper from us at @MSKCancerCenter explores what happens after we *all* have been infected
biorxiv.org/content/10.64898… 1/n
About 90% of RNA molecules in our cells are tRNAs. To date, we have mostly been blind to their abundance and regulation in single-cell sequencing. Today, I’m excited to share a new manuscript from my MD/PhD work
biorxiv.org/content/10.64898…
If you are already using Flex, quantifying tRNAs is a straightforward and modular addition. Please reach out if you want to learn more. I’d be excited to discuss tRNAs and other science with whoever has thoughts on our manuscript and beyond!
1
5
448




