Nick Venanzi

Science is entering a new era - one where AI agents can do scientific work. 🧬 Today NVIDIA is launching the BioNeMo Agent Toolkit - an open, agent-ready toolkit that gives any AI agent callable tools for protein structure prediction, molecular docking, generative chemistry, genomic analysis, and more. (1/2)

In collaboration with @PeptoneLtd, NVIDIA researchers are pushing the boundaries of protein science through AI. Their latest work explores disordered proteins and introduces two major innovations: - PeptoneBench: The first benchmark designed specifically for disordered protein ensembles, integrating data from NMR, SAXS, PRE, and RDC experiments - PepTron: A powerful BioNeMo-based generator that leverages flow matching, cuEquivariance, and Megatron parallelism to model disordered proteins at scale Learn more ⬇️

@DrGlycoBen @LabMcDonald @MerryLabNotts @lwalport Congrats 🎉🎉

For ml4lms.bio we are in need for some more reviewers in: - computer vision applied to life sciences and medical imaging - material sciences - population healthcare - life sciences Consider filling out forms.gle/g2RYNCKTAPPsBj… with your OpenReview email to be added!

📢 Check the new Volume 64 Issue 7 of #JCIM pubs.acs.org/toc/jcisd8/64/7

❗️ Submissions are open now until until May 6th ❗️ Instructions here: ml4lms.bio/submission

Machine Learning Integrating Protein Structure, Sequence, and Dynamics to Predict the Enzyme Activity of Bovine Enterokinase Variants #machinelearning #compchem pubs.acs.org/doi/abs/10.102…

@KevinKaichuang Wow me too actually 😅

I got into ml because the other option was organic chemistry

@KevinKaichuang Super cool !

EvoDiff combines evolutionary-scale data with diffusion models for controllable protein sequence generation. In addition to generating plausible proteins, we can scaffold structural motifs in sequence space! Preprint: biorxiv.org/cgi/content/sh… Code: github.com/microsoft/evod…

@sokrypton I can confirm you are not alone haha

Am I the only one that needs to google how to use the python pickle library every-single-time? 😅

@sokrypton @MITBiology How to join ? ;)

I'm excited to share that I'll be joining @MITBiology as an Asst Prof. in Jan 2024! Come join us! 🤓🧪🖥️🧬

@andrewwhite01 @gpt_index Loving this !

I've been working on improving LLM's ability to cite papers with the following chain: 1. Convert question into search query for google scholar 2. Scan PDFs of returned papers and embed into @gpt_index 3. Extract context for question and present in prompt. Expensive, but amazing

@simonduerr Was quite a lesson on overfitting I would say ! Quite interesting work and approaches have been shown! More of bio protein comps please 😀

Results for this competition are online since a few days. Interesting that the very good intermediate results (Rs=0.8+) were mainly due to people overfitting to the public test data 🥸. Final winner achieved Rs ~0.55 for the private test data. 1/2

@KevinKaichuang @fried_lab @BookerMan_PSU The link is to proteinGAN

@nathanbenaich About damn time!

Overheard from an ML engineer in self-driving: "I can see a lot of brain drain happening to biotech" IT IS WORKING!

@amywang01 @workshopmlsb @avapamini @alexijielu @KevinKaichuang Looks super!

I'll be presenting a poster on learning from physics-based features to improve protein property prediction at @workshopmlsb. This is work from an MSR internship, where I got to work with amazing people like @avapamini @alexijielu and @KevinKaichuang!

Super excited to be at NeurIPS tomorrow and Sat (12/2-12/3)! Would love to meet up and chat about biophysics+ML 🥳

Hi Tweeter! I am looking for a computational person in the bio-world (proteins) to help me build a platform that uses physics and AI. Please do get in touch if this pokes your interest!😀

@DrGlycoBen @CarolynBertozzi Yes yes yes!!

@JoshuaB50143787 What strat is this ? :)

Etherum Merge #Etherum #ETH $ETH

@KevinKaichuang This is so cool!

Use a neural network to predict binding affinity from MD simulations. Yaosen Min, Ye Wei, Peizhuo Wang, Nian Wu, Stefan Bauer, Shuxin Zheng, Yu Shi, Yingheng Wang, Dan Zhao, Ji Wu, Jianyang Zeng arxiv.org/abs/2208.10230