Training data for conformational flexibility prediction
How much data from molecular dynamics simulations are needed to predict protein flexibility?
»Antibody design de novo vs in vivo
The Baker lab tackles de novo antibody design by narrowing the problem as much as possible.
»Low-dimensional representations of MD simulations
Autoencoders, a type of neural network that learns how to optimally compress information, share some superficial resemblances to collective variables (CVs) used in MD simulations.
»Structural adaptors for protein property prediction
It’s always tricky to choose which protein neural network to use for fine-tuning tasks.
»Estimating an antibody's conformational landscape using inverse folding
Calculating the full conformational landscape of an antibody, or any medium-sized protein more generally, is computationally expensive. A new preprint introduces a shortcut that could speed this up and accelerate antibody design.
»Observations on how antibodies evolve
A recent study on anti-SARS-CoV-2 antibodies retraces the steps they take during affinity maturation.
»Nanobody CDR3 structure prediction
Four years after the protein folding problem was allegedly solved, we still can’t reliably predict how or where antibodies bind to their antigens. A recent report identifies one source of continued difficulty.
»