Analyze and design proteins with ChatMol copilot! Unaccelerated demo video: open source code: Here is an accelerated recording.

The prompt to create the image is: "show 8RWR and render it in a creative way with protein-ligand details. remove waters." some mirror changes are need to get the final image.

📢 Just updated ChatMol with some major improvements: 🔌 Switched to direct API calls - no more OpenAI SDK dependency 🤖 Added Anthropic API support to use Claude models 🏗️ Cleaned up the codebase architecture Try it out: #PyMOL #AI #Agent #LLM

Try it now: #BioinfoTools #MachineLearning #ProteinEngineering

RT @pengzhangzhi1: By the way, @jinyuansun39143 just implemented Progen2 with Flash Attention. Save up to 60% of memory and run time by us…

pLM are used everywhere is biology today. I provide a "Simple, unified interface to multiple protein language models", called plmsuite at:

RT @biorxiv_bioinfo: Language model generates cis-regulatory elements across prokaryotes #biorxiv_bioinfo

@KevinKaichuang I was so interested in this one but my mentor did not like this. Guess I should be your intern back to my first year of my Phd. lol

@DdelAlamo This make me sad, I was so excited to see the generative model.

@KevinKaichuang @KieraSumida @DaveJuergens finally find out the ChemNet used in RFAA paper for docking analysis...... 😂

@AjasjaLjubetic Can not wait to try it. Setting up rosetta and pyrosetta in new machines is still unplesent after all these years using it.

I gues af3 is promising in sampling conformaitonal changes. use the helixfold3 weights and code, I was able to easily find fs states of KaiB. @sokrypton I would like any suggestion from you.

@RolandDunbrack @sokrypton @DdelAlamo @GoogleDeepMind I guess they will. Baidu made a helixfold 3 and claimed to reproduce the af3. Unless deepmind largely improved the af3 server's prediction, there is no reason to not release the af3 model.

@charlieharris01 Thanks! We are trying the best to make ChatMol better!

RT @charlieharris01: ChatGPT in PyMol...the future really has arrived

Dose anyone know about the --seq_limit flag? This appeared in the AF3 paper but without any explanation. #alphafold

RT @pacyc1841: Species-specific design of artificial promoters by transfer-learning based generative deep-learning model | Nucleic Acids Re…

RT @NatureComms: Computationally engineered TurboPETase achieves nearly complete depolymerization of postconsumer PET bottles at a high ind…