De novo design of macrocyclic molecular glues with EvoBind-multimer
EvoBind-multimer (EBM) designs novel linear and cyclic peptides bridging two target proteins, only from their amino acid sequences. This makes it possible to design peptide molecular glues that create ternary complexes. It is not necessary to specify any target residues within the protein sequence or the length of the peptide (although this is possible).
We demonstrate molecular glue design with experimental validation between an E3 ligase and a target protein for the degradation of the "undruggable" oncoproteins KRAS and BRD4.
Read more here
EvoBind-multimer accounts for adaptation of the receptor interface structures to the peptide being designed during optimization: sequence and structure is generated simultaneously. This consideration of flexibility is crucial for binding. EvoBind-multimer is built on the ideas of EvoBind2, the first protocol that only relies on a protein sequence to design a binder with experimentally verified cyclic design capacity.
The animation illustrates the real-time design trajectory: the target protein KRAS (grey) and the E3 ligase VHL (green) dynamically adapt their interface conformation as the cyclic peptide molecular glue (blue) sequence and structure are generated simultaneously.
- EvoBind-multimer
- LICENSE
- Colab
- Computational requirements
- Setup
- Design molecular glues
- Citation
- The EvoBind ecosystem
EvoBind-multimer is based on AlphaFold2, which is available under the Apache License, Version 2.0.
The AlphaFold2 parameters are made available under the terms of the CC BY 4.0 license and have not been modified.
The design protocol EvoBind-multimer is made available under the terms of the CC BY-NC 4.0 license.
You may not use these files except in compliance with the licenses.
It is possible to run EvoBind-multimer online in the Google colab here
Before beginning the process of setting up this pipeline on your local system, make sure you have adequate computational resources. Make sure you have an available GPU as this will speed up the prediction process substantially compared to using a CPU. EvoBind-multimer assumes you have NVIDIA GPUs on your system, readily available. A Linux-based system is assumed.
To setup this pipeline, clone this github repository:
git clone https://github.com/patrickbryant1/EvoBind-multimer.git
Then do
cd EvoBind-multimer
bash setup.sh
This script fetches the AlphaFold2 parameters, installs a conda env and downloads uniclust30_2018_08 which is used to generate the receptor MSA.
To design binders the following needs to be specified:
FASTA1: protein target 1 amino acid sequence as .fasta
FASTA2: protein target 2 amino acid sequence as .fasta
Optional arguments:
Peptide length - default=10
Target residues within the target 1 and/or target 2 - default=all
If you want to design a cyclic peptide, add the flag --cyclic_offset=1 in the design script when calling mc_design.py. Based on cyclic offset.
A test case is provided in design_local.sh.
This script can be run by simply doing:
bash design_local.sh
If you use EvoBind-multimer in your research, please cite
De novo Design of Macrocyclic Molecular Glues Andra Brunner, Krzysztof Wierbilowicz, Diandra Daumiller, Daniel Bexell, Kasper Karlsson, Olle Sangfelt, Patrick Bryant bioRxiv 2026.08.21.746227; doi: https://doi.org/10.64898/2026.08.21.746227
EvoBind - designs novel [cyclic] peptide binders based only on a protein target sequence.
RareFold - prediction & design with noncanonical amino acids
RareFoldGPCR - GPCR agonist design with noncanonical amino acids