Aakarsh Vermani
@aakarshv1
@UCBerkeley EECS | @BerkeleyML | Research @ CCB
ID: 1262089293444444160
17-05-2020 18:35:27
11 Tweet
130 Followers
192 Following
Come hear Aakarsh Vermani talk about machine learning for protein engineering! The talk will give an overview of AlphaFold2, RFDiffusion, and ProteinMPNN, three models that have been instrumental in the areas of protein structure prediction, and design. lu.ma/4baniw2s
A friend sent me this great vid by N Shackleton-Jones about how routine makes lives feel shorter, an idea backed by psych research. I think there’s a good info theory perspective here: routine = low entropy so your memories can be encoded more efficiently and thus feel shorter!
Its been a blast setting up and organizing the BioML seminar this year w/ Samarth Jajoo stacy 🌤! We've uploaded the recorded seminars with Sam Rodriques, Das Lab and Aakarsh Vermani's excellent primer on ML methods in protein engineering! Check 'em out: youtube.com/watch?v=opeKuk…
Antibodies are highly diverse, but most possible sequences are unstable or polyreactive. In this work led by Milind Jagota, we propose a new source of data for modeling constraints from these properties. Our models show clear improvements in predicting antibody dysfunction.(1/n)
Join us on Monday 3/10 for our latest installment of the BioML @ Berkeley seminar series! We'll be learning from the exceptional Elana Simon (Elana Simon) about mechanistic interpretability in BioML. lu.ma/guiyjbf9
We're having another BioML @ Berkeley seminar next Monday 4/7! This time we'll be learning from David Kelley (David Kelley) about Borzoi, a new model that predicts RNA-seq coverage from DNA sequence, enabling a better understanding of gene regulation. lu.ma/yjk0a1us