mcguffin/multifold2 repository overview

⁠MultiFOLD2, ModFOLDdock2 and MultiFOLD2_refine: prediction, quality assessment and refinement of protein tertiary and quaternary structure models

A docker container for the MultiFOLD2, ModFOLDdock2 and MultiFOLD2_refine methods developed by the McGuffin group⁠ for CASP16⁠.

⁠Please cite the relevant papers:

MultiFOLD2 and ModFOLDdock2: McGuffin, L. J., Alhaddad, S. N., Behzadi, B., Edmunds, N. S., Genc, A. G., and Adiyaman, R. (2025) Prediction and quality assessment of protein quaternary structure models using the MultiFOLD2 and ModFOLDdock2 servers. Nucleic Acids Research, gkaf336. DOI⁠ PubMed⁠

MultiFOLD: McGuffin, L. J., Edmunds N. S., Genc, A. G., Alharbi, S. M. A., Salehe, B. R. and Adiyaman, R. (2023) Prediction of protein structures, functions and interactions using the IntFOLD7, MultiFOLD and ModFOLDdock servers. Nucleic Acids Research. gkad297. DOI⁠ PubMed⁠

ModFOLDdock: Edmunds, N. S., Alharbi, S. M. A., Genc, A. G., Adiyaman, R. and McGuffin, L. J. (2023) Estimation of Model Accuracy in CASP15 Using the ModFOLDdock Server. Proteins, Epub ahead of print. DOI⁠ PubMed⁠

MultiFOLD_refine: Adiyaman, R., Edmunds, N. S., Genc, A. G., Alharbi, S. M. A. and McGuffin, L. J. (2023) Improvement of protein tertiary and quaternary structure predictions using the ReFOLD refinement method and the AlphaFold2 recycling process. Bioinformatics Advances, vbad078. DOI⁠

Note: For ModFOLDdock2Q, the lightweight version of ModFOLDdock2, click here⁠

⁠Requirements

The latest version of Docker and the NVIDIA Container Toolkit⁠.

NOTE: This version has been tested to work on Ubuntu 22.04 with NVIDIA L40S GPUs.

⁠Setup

Optionally set the input and output paths in your shell (alternatively just include the full paths when you run):

export MF2_IN=/location/for/your/input/fasta_files/

export MF2_OUT=/location/for/your/output/files/

(Note: The commands below can be also be run without sudo by following the Docker Engine post-installation steps⁠.)

⁠Run standard MultiFOLD2 for 3D modelling of multimers from sequences in FASTA format

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_MultiFOLD2_docker.sh targetname fastafile stoichiometry

e.g. Run MultiFOLD2 for CASP16 target "H0232" with the FASTA file "H0232.fasta⁠" and stoichiometry "A2" as follows:

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_MultiFOLD2_docker.sh H0232 H0232.fasta A2B2

Note: regarding the stochiometry parameter, a monomer=A1, a homodimer=A2, a homotrimer=A3, etc. A heterodimer=A1B1, a heterotrimer=A1B1C1, etc. You can also model targets with more complex stoichiometries, such as a dimer of dimers=A2B2, a trimer and dimer=A3B2, a nonamer and trimer=A9B3, and so on. You must provide a sequence in the FASTA file for each subunit in your complex. If the stoichiometry is not known, then you can use the parameter "unknown" and it will be predicted from templates, e.g.,

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD_input -v $MF2_OUT:/MultiFOLD_output mcguffin/multifold /MultiFOLD/run_MultiFOLD_docker.sh H0232 H0232.fasta unknown

MultiFOLD2 output: The output for each job will be found in $MF2_OUT within a unique subdirectory named according to the time and date of submission plus the extension "_TS". Your final results file will be in CASP TS format and can be found within the "MultiFOLD2_TS_" subdirectory with a filename containing the unique sequence ID for your job and ending with the extension "_TS.txt" e.g.

$MF2_OUT/14_52_44_883_18-1-2025_TS/MultiFOLD2_TS_r28d3fk5p89l3fvc/r28d3fk5p89l3fvc_TS.txt

All 3D models in PDB format can be found within the MultiFOLD_models/ subdirectory e.g.

$MF2_OUT/14_52_44_883_18-1-2025_TS/MultiFOLD2_TS_r28d3fk5p89l3fvc/MultiFOLD2_models/

The final top 5 models and ranking data can be found in the final_TS_data/ subdirectory e.g.

$MF2_OUT/14_52_44_883_18-1-2025_TS/final_TS_data/

⁠Run MultiFOLD2_refine for refining 3D models of complexes with known stoichiometry

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_MultiFOLD2_refine_docker.sh targetname fastafile modelfile stoichiometry recycles

e.g. Run MultiFOLD2_refine for CASP15 target "H1232" with the FASTA file "H1232.fasta⁠" on the model "H1232TS163_1.pdb⁠" with stoichiometry "A2B2" for "12" recycles as follows:

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_MultiFOLD2_refine_docker.sh H1232 H1232.fasta H1232TS163_1.pdb A2B2 12

MultiFOLD2_refine output: The output for each job will be found in $MF2_OUT within a unique subdirectory named according to the time and date of submission plus the extension "_REF". Your final results file will be in CASP TS format and can be found within the "MultiFOLD2_TS_" subdirectory with a filename containing the unique sequence ID for your job and ending with the extension "_TS.txt" e.g.

$MF2_OUT/14_52_44_883_18-1-2025_REF/MultiFOLD2_TS_r28d3fk5p89l3fvc/r28d3fk5p89l3fvc_TS.txt

All 3D models in PDB format can be found within the MultiFOLD2_models/ subdirectory e.g.

$MF2_OUT/14_52_44_883_18-1-2025_REF/MultiFOLD2_TS_r28d3fk5p89l3fvc/MultiFOLD2_models/

⁠Run ModFOLDdock2 for estimating the quality of 3D models of complexes

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_ModFOLDdock2_docker.sh targetname fastafile modelfile stoichiometry

For estimating the quality of a single model:

e.g. Run ModFOLDdock for CASP15 target "H1232" with the FASTA file "H1232.fasta⁠" on the single model "H1232TS163_1.pdb⁠" with stoichiometry "A2B2" as follows:

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_ModFOLDdock2_docker.sh H1232 H1232.fasta H1232TS163_1o.pdb A2B2

For estimating the quality of multiple models:

e.g. Run ModFOLDdock for CASP15 target "H1232" with the FASTA file "H1232.fasta⁠" on multiple models "H1232.tar.gz⁠" with stoichiometry "A2B2" as follows:

sudo docker run --gpus all -v $MF2_IN:/MultiFOLD2_input -v $MF2_OUT:/MultiFOLD2_output mcguffin/multifold2 /MultiFOLD2/run_ModFOLDdock2_docker.sh H1232 H1232.fasta H1232.tar.gz A2B2

ModFOLDdock2 output: The output for each job will be found in $MF2_OUT within a unique subdirectory named according to the time and date of submission plus the extension "_QA". Your final results file will be in CASP15 QA format and can be found within this subdirectory with a filename containing the unique sequence ID for your job and ending with the extensions: "_QA.txt" for standard ModFOLDdock2 jobs, "_R_QA.txt" for ModFOLDdock2R jobs, or "_S_QA.txt" for ModFOLDdock2S jobs e.g.

$MF2_OUT/14_52_44_883_18-1-2025_QA/r28d3fk5p89l3fvc_QA.txt

$MF2_OUT/14_52_44_883_18-1-2025_QA/r28d3fk5p89l3fvc_R_QA.txt,

or

$MF2_OUT/14_52_44_883_18-1-2025_QA/r28d3fk5p89l3fvc_S_QA.txt

⁠Contact

Prof. Liam J. McGuffin⁠

⁠License

MIT License

Copyright (c) 2025 Liam McGuffin, Recep Adiyaman, and the University of Reading

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Third-Party Software Notices for MultiFOLD2 Docker image⁠