https://youtu.be/yqeUH4RsJp8?si=SkolqnGxO96VVaVI

Amazingly clear lectures for understanding Machine Learning protein structure prediction!

Part I - Lucid overview of why different ML algos map best to bio structures

Part II - Protein 3D structure boils down to well-informed placement of backbone triangles, from amino acid co-evolution

Part III - Where to now for systems biology? We've gone gene -> network but skipped the machines

Will be exciting when chemical interaction constraints are incorporated!

#OpenFold

Nazim Bouatta | Machine learning for protein structure prediction, Part 1: Algorithm space

YouTube

@xtaldave @strucbio

OpenFold is a big deal. And a great showcase for free open source software, enabling scientists to improve on each other's work in the best scientific tradition.

Source code: https://github.com/aqlaboratory/openfold

Open data for training: https://registry.opendata.aws/openfold/

#OpenFold #FOSS #StructuralBiology #CryoEM

GitHub - aqlaboratory/openfold: Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2

Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2 - aqlaboratory/openfold

GitHub
What a week... #OpenFold is out and supposedly faster and more memory efficient then #AF2. This reaches the #AF3 token limit on a local GPU 👍 (only with proteins though...) #StructuralBiology @strucbio https://www.nature.com/articles/s41592-024-02272-z?utm_source=twitter&utm_medium=social&utm_campaign=nmeth
OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization - Nature Methods

OpenFold is a trainable open-source implementation of AlphaFold2. It is fast and memory efficient, and the code and training data are available under a permissive license.

Nature

Currently testing Folding Playground

A one-step solution for comparing #alphafold #omegafold #esmfold and #openfold #AI #protein #structure

https://www.sphinxbio.com/post/folding-playground

Sphinx Bio | The modern biotech data platform

Reading #AlphaFold2 and #OMSF #OpenFold code is great but with #protein #molecules in your hands you feel it better. Designing the kit in #FreeCAD ensures you know bond lengths and angles etc. So I spent few days learning and printing. The kit is optimized for protein #folding, single bond rotates but can be fixed in #peptide bond, it keeps shape even before adding #hydrogen #bonds in #alphahelix (or pair #RNA bases by H-bond).

Print your #kit and enjoy #biochemistry!

https://www.printables.com/model/457984-molecules-for-protein-folding

Printables

What is your #openfold setup?
What #GPU do you use?
Are you happy with it?