Twitter | Pretraživanje | |
Surge Biswas
Interested in what machine learning can do for protein engineering & sports. lab . PhD candidate . Dog dad.
306
Tweetovi
340
Pratim
605
Osobe koje vas prate
Tweetovi
Surge Biswas 30. sij
🤯Using this to calculate tips from now on
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Jasper 29. sij
Cool new result in protein design. Multitask from large data to small - transfer learning from a general protein representation to one tailored to a particular function of interest.
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
PlesaLab 28. sij
We're hiring a bioinformatician to help us tackle a number of projects involving large scale sequence design, optimization, and analysis:
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Grigory Khimulya 28. sij
In DC for , so much cool work here in rapid countermeasure response for new pathogens! If you are working on this, I want to talk - dms open.
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Kevin Yang 楊凱筌 27. sij
Great work on using ML for protein engineering. Key insights: - fine-tuning global LM using local landscape helps - pretrained model can predict epistatic effects from single mutants
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 25. sij
Odgovor korisniku/ci @giessel
Thanks for reading Andrew! Surprising for us too! Definitely curious to hear more thoughts on that finding if you have them
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
andrew giessel 25. sij
Odgovor korisniku/ci @SurgeBiswas
This was fantastic! The PC1 correlation is actually just amazing.
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 25. sij
Odgovor korisniku/ci @fischer_cr @grigonomics i 3 ostali
haha. or less asparagine?
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Gleb Kuznetsov 24. sij
Interesting potential for ML-guided protein engineering: might be able to systematically optimize proteins after just a few choice measurements. Check out this new preprint from and team.
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Jase Gehring 24. sij
won't stop me from working on ways to get big-N, but sometimes small-N is hard enough
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Anush Chiappino-Pepe 25. sij
Super cool work from , , , , and ! Efficiently identifying mutant proteins with higher activity using ML, i.e., eUniRep. Congrats!!!
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Mohammed AlQuraishi 24. sij
Had an early look at this work and it’s really impressive stuff! Demonstrates the remarkable power of semi-supervised learning in very low N contexts.
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas proslijedio/la je tweet
Ethan C. Alley 24. sij
Our new paper! My favorite bits were: - Discovering WAY more functional proteins out there (1000+!!) then previously explored by evolution OR decades of engineering - Connecting engineering and tech-translation failures with Goodhart's law - 's first thread :P
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 24. sij
Odgovor korisniku/ci @SurgeBiswas
Hope you enjoy the read! We haven't submitted this anywhere yet so feedback and venue/journal suggestions welcome! DMs open
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 24. sij
Odgovor korisniku/ci @grigonomics @EthanAlley i 2 ostali
As always, inspiring to work with and learn from and , and have the mentorship of and :)
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 24. sij
Odgovor korisniku/ci @SurgeBiswas
For the ML folks reading this, we're excited by the extreme data-efficiency we saw here, and for forward design of a complicated natural object no less. Another feather in the cap of unsupervised/semi-supervised learning, no doubt!
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 24. sij
This has been a really fun one! challenging the wisdom that one needs a lot of data to do ML guided biodesign. In fact, if you have to choose, think about getting HQ small-N data aligned w your endpoint, vs HT proxy data. low-N ML now helps you succeed w the former!
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 19. sij
*Turns off FACS machine*
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 17. sij
11/10 opportunity to work for a great company, , with an insanely fun and smart person!
Reply Retweet Označi sa "sviđa mi se"
Surge Biswas 16. sij
Odgovor korisniku/ci @ebenbayer
Is there a faster form of info transfer you prefer?
Reply Retweet Označi sa "sviđa mi se"