New Title Alert: mBER-open- is a protein design framework for antibody binder design that leverages structure templates and sequence conditioning in backpropagation-based design through AlphaFold-Multimer.
Learn more here: https://github.com/manifoldbio/mber-open/
New Title Alert: easymode- a collection of general pretrained networks for cellular cryoET.
Learn more here: https://mgflast.github.io/easymode/
New Title Alert: ThermoMPNN- is a graph neural network (GNN) designed to predict changes in stability for protein point mutants using transfer learning.
Learn more here: https://github.com/Kuhlman-Lab/ThermoMPNN
New Title Alert: Slabify- is a command-line tool designed for the automatic segmentation of lamella slabs in cryo-electron tomography (cryo-ET) volumes.
Learn more here: https://github.com/CellArchLab/slabify-et
New Title Alert: MagCalEM- Accurate magnification determination for cryoEM determining the correct magnified pixel size of single-particle cryoEM micrographs is necessary to maximise resolution and enable accurate model building.
Learn more here: https://www.mrc-lmb.cam.ac.uk/crusso/magCalEM/index.html
New Title Alert: LASErMPNN- is an all-atom ligand-conditioned protein sequence design and sidechain packing model that accounts for the presence of small molecules, including hydrogens.
Learn more here: https://github.com/polizzilab/LASErMPNN/tree/main
New Title Alert: IsoNet2- a deep-learning software package for simultaneous missing wedge correction, denoising, and CTF correction in cryo-electron tomography reconstructions using a deep neural network trained on information from the original tomogram(s).
Learn more here: https://github.com/IsoNet-cryoET/IsoNet2
Technical Notes: January's software push includes updates to 15 titles -- anvi’o, BioXTAS RAW, ChemEx, CryomaskR, DeepRank-GNN-esm, DSSR, GROMACS, Icecream, IMP, PowerFit, PyEM, RDKit, SRAToolkit, Volume_Seg_Tool, and XDS.
More here: https://sbgrid.org/software/recent/archive/2026/01/
New Title Alert: ImmuneBuilder- is a high-performance deep learning framework developed by the Oxford Protein Informatics Group (OPIG) specifically for predicting the 3D structures of immune receptor proteins.
Learn more here: https://github.com/oxpig/ImmuneBuilder
New Title Alert: Foundry- is a unified framework for protein design that provides tooling and infrastructure for using and training models including RFdiffusion3 (all-atom generative design), RosettaFold3 (structure prediction), and ProteinMPNN/LigandMPNN (inverse folding).
Learn more here: https://rosettacommons.github.io/foundry/
SBGrid
