For those of you at @Winter_QBio: tomorrow morning, @AndreNguyen16 presents our new research on estimating the effects of mutations outside the #CDR3 region on #TCR-#epitope binding in the context of primary human T cells. Puerto Rico Ballroom 5 at 8:40 a.m Wednesday! 1/n
I have never been prouder of our #ML pipeline than these past few weeks as multiyear projects have begun to converge: now we have more experiments to run. Is anyone with a background in immunology #wetlab experience looking for #researchscientist roles at the moment? #Boston area
Oooh! Our work on determining contributions to binding outside of the CDR3 (and inside) has been accepted for a contributed talk @Winter_QBio
Hurray @EliWeinstein6@AndreNguyen16!!
Anyone headed there in person?
For the record, the other people on the @lmrl_bio organizing committee have gone above and beyond this whole year, as have those that have submitted and the dozens that helped review. But the first we heard about the 11/25 deadline was on 11/21. Help us, @sanmikoyejo@shakir_za!
Oh god. I went on maternity leave, was the main P.O.C., was 2 days late to upload our schedule (deadline was 11/25), and now @lmrl_bio has been cancelled by #NeurIPS2022 (seriously.) Please send help.🫠
The first Human Immunome AI Summit is in full swing. World leading scientists across #immunology, computational #biology, #AI are developing a strategic roadmap for how to build the first AI model of the human #immunome, capable of leading science into #digital age. #HIPAISummit
Thrilled: @lmrl_bio was accepted for @NeurIPSConf 2022. For lots of reasons, we'll be virtual -- hope that makes it easy and safe for you to attend! Pre-register for workshop updates at https://t.co/bH7HENF2mP and join the Slack, follow on Twitter, etc. Thanks for sharing!
The preprint of my PhD work w/ Jeff Miller is now up. In it we study the problem of "data selection", finding low-dimensional statistics of high-dimensional data that can be fit by a parametric model of interest. https://t.co/d988l6Sb7W
Our work exploring "the use of neural networks for influenza prediction at multiple spatial resolutions" is now out https://t.co/4rHYJ4qDME Great collaboration with E Aiken, @AndreNguyen16 and C Viboud
An early warning approach to monitor COVID-19 activity with multiple digital traces in near real time. Great collaboration with an outstanding scientific team https://t.co/8e2TmwZbf2
Coauthors include: @nkiocgoalne@SKNKsmash P Liautaud, J Kaashoek, @NicholasLink10 ...
So big personal news, I've been working on a book w/ @ManningBooks and it's now early-access! https://t.co/4gCsOCf4p7 I'm targeting what I think is an under-served area. The middle between "give me a tool" and "CS/Stats/ML PhD graduate book" that gives utility and understanding.
We're so excited to have SWA natively in PyTorch 1.6! Improved generalization, semi-supervised learning, deep RL, uncertainty, Bayesian DL, calibration, low precision training... I encourage you try it out! It's easy, and also works on pre-trained models. https://t.co/LJ71NzeU6p
Can we forecast the near future of the pandemic with data we have now? That’s the goal of this work led by @MauSantillana. The idea is simple;🔼in pandemic activity will lead to changes in multiple data streams before increases in cases become apparent https://t.co/WvJDsV8IbB 1/n