Anthony Gitter
@anthonygitter
Followers
3K
Following
8K
Media
271
Statuses
3K
Computational biologist; Associate Prof. at University of Wisconsin-Madison; Jeanne M. Rowe Chair at Morgridge Institute
Joined April 2015
OpenFold3-preview (OF3p) is out: a sneak peek of our AF3-based structure prediction model. Our aim for OF3 is full AF3-parity for every modality. We now believe we have a clear path towards this goal and are releasing OF3p to enable building in the OF3 ecosystem. More👇
2
61
197
ColabSaprot & SaprotHub are now in @NatureBiotech! 🧬 A user-friendly, no-code platform for training, sharing, and collaborating on protein language models. We also provide ColabSeprot, integrating ESM1b, ESM2, ProTrek, and ProtBert for the community. https://t.co/OIh5t7uvPE
10
41
146
The next Adaptyv Bio protein design competition will select 200 of its 1000 designs by community vote. Tell all your friends and see if your name will bind Nipah virus Glycoprotein G!
🚨 The Protein Design Competition is back: Bigger, better, and more impactful than ever. This time, the global protein design community will test 1,000 novel protein sequences against one of the deadliest known viruses: 🦠Nipah. Last year, we challenged the community to design
0
1
3
🚨 The Protein Design Competition is back: Bigger, better, and more impactful than ever. This time, the global protein design community will test 1,000 novel protein sequences against one of the deadliest known viruses: 🦠Nipah. Last year, we challenged the community to design
🌍 The biggest decentralized science experiment of 2025 is starting now! The protein design competition returns: we’re inviting scientists, engineers, and hackers from around the world to help design new proteins capable of neutralizing the Nipah virus, a pathogen with up to 75%
1
10
18
I felt a great disturbance in the Scientific Agents, as if millions of startups suddenly cried out in terror and were suddenly silenced.
We’re building tools to support research in the life sciences, from early discovery through to commercialization. With Claude for Life Sciences, we’ve added connectors to scientific tools, Skills, and new partnerships to make Claude more useful for scientific work.
5
2
134
LLM engineered carbon capture enzymes have officially been produced. The best designs were 170% more active and 25% more stable across extreme pH (Tm +8.5 C). Winning strategies include adapting a tag from a bacterial carbonic anhydrase, beta barrel core packing, and removing
31
80
431
I decided to experiment with my own (small) protein design competition! It's specifically to test how well VHH pipelines work for binder design, without the usual careful tuning. You submit a @modal script, I run it and test on https://t.co/z5f9FrA7qJ
https://t.co/7NGUw4duVH
1
17
63
I received many requests to share materials from our undergraduate course “Machine Learning in Chemistry” — here you go! A preprint summarizing insights and lessons learned: https://t.co/UcQbWet75n A Jupyter Notebook Tutorial Gallery: https://t.co/dcgBzsTTe6
My focus for Spring 2025: launching an undergraduate course @UWMadisonChem @TCI_UW_Madison developed from scratch - "Chem361: Machine Learning in Chemistry"! Here's a glimpse of what we'll explore:
7
92
466
The journal version of our Multi-omic Pathway Analysis of Cells (MPAC) software is now out: https://t.co/HBqN0DerzL MPAC uses biological pathway graphs to model DNA copy number and gene expression changes and infer activity states of all pathway members.
academic.oup.com
AbstractMotivation. Fully capturing cellular state requires examining genomic, epigenomic, transcriptomic, proteomic, and other assays for a biological sam
Our Multi-omic Pathway Analysis of Cancer (MPAC) software is now available on bioRxiv and Bioconductor. MPAC uses biological pathway graphs to reason about DNA copy number and gene expression changes and infer activity states of all pathway members.
0
0
3
It's great to see @proteinbase launch! I really like the interface and shared my Adaptyv data there during the beta test, which is why you'll see Llama 3.1 405B and BLOSUM62 listed as protein design methods.
Today we’re releasing real-world experimental data for over 1000 novel AI-designed proteins on our new platform @proteinbase!
0
2
13
New blogpost on the latest in AI antibody design. Including some code to easily run Germinal and IgGM on modal! https://t.co/G2ZC6fzUxq
11
50
196
Evaluation of machine learning-assisted directed evolution across diverse combinatorial landscapes 1. Machine learning-assisted directed evolution (MLDE) strategies have been shown to be more efficient than traditional directed evolution methods in identifying high-fitness
0
13
74
Big update on our Pinal preprint! We report four successful experimental validations for our designed proteins: PETase, GFP, ADH, and H protein.🧬 Notably, our designed H protein exhibits a 1.7x performance boost over its natural counterpart. 🚀 Try: https://t.co/4XxGI0Hp9x
5
21
117
Mutational effect transfer learning (METL) is a biophysics-based protein language model that excels in challenging protein engineering tasks. https://t.co/NTUrz8IuJe
0
4
31
The main GitHub repo https://t.co/PdLzaT2kWR links to the extensive resources for running Rosetta simulations at scale to generate new training data, training METL models, running our models, and accessing our datasets. Paper:
nature.com
Nature Methods - Mutational effect transfer learning (METL) is a protein language model framework that unites machine learning and biophysical modeling. Transformer-based neural networks are...
0
2
7
The journal version of our "Biophysics-based protein language models for protein engineering" with @romerolab1 is now live! We train a protein language model on biophysical simulations from @RosettaCommons to support protein engineering.
Our manuscript "Biophysics-based protein language models for protein engineering" with @romerolab1 is now on bioRxiv. We present Mutational Effect Transfer Learning (METL), a protein language model trained on biophysical simulations, and showcase it for protein engineering. 1/
3
16
133
Latest genomic AI models report near-perfect prediction of pathogenic variants (e.g. AUROC>0.97 for Evo2). We ran extensive independent evals and found these figures are true, but very misleading. A breakdown of our new preprint: 🧵
9
118
486
However, an improvement to BioML-bench would be to only use certified @Polaris_HQ datasets. Some uncertified datasets like tdcommons-bbb-martins have known problems.
0
0
0
As more and more AI agents for biology appear, I hope to see new benchmarking efforts like BioML-bench https://t.co/hnOnoGrJQc It includes important metrics like completion rate.
2
1
10
🚨The Bits to Binders Competition has concluded!🧬 One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors Spoiler: They work!
4
42
252