https://github.com/jhrcook/proof-of-concept-alphafold
A proof-of-concept for using AlphaFold with new amino acid sequences (this may change to a full project in the future).
https://github.com/jhrcook/proof-of-concept-alphafold
alphafold bioinformatics jupyterlab protein-folding python3 tensorflow
Last synced: 8 months ago
JSON representation
A proof-of-concept for using AlphaFold with new amino acid sequences (this may change to a full project in the future).
- Host: GitHub
- URL: https://github.com/jhrcook/proof-of-concept-alphafold
- Owner: jhrcook
- Created: 2020-12-02T14:39:43.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2020-12-02T14:40:12.000Z (almost 5 years ago)
- Last Synced: 2025-01-13T11:50:09.518Z (9 months ago)
- Topics: alphafold, bioinformatics, jupyterlab, protein-folding, python3, tensorflow
- Language: Shell
- Homepage:
- Size: 6.84 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Proof-of-Concept for using DeepMind's AlphaFold model
Feature generation using the scripts from [Urinx/alphafold_pytorch](https://github.com/Urinx/alphafold_pytorch).
Predictions using the original [AlphaFold](https://github.com/deepmind/deepmind-research/blob/master/alphafold_casp13/requirements.txt) model.
## Set-up
### Feature creation
#### Conda virtual environment
I followed the instructions in [Urinx/alphafold_pytorch](https://github.com/Urinx/alphafold_pytorch) for setting up the feature creation workflow.
To create the `conda` virtual environment and install most of the required packages, run the `setup.sh` script.
The `plmDCA` tools needs to be setup manually following the instructions in [Urinx/alphafold_pytorch](https://github.com/Urinx/alphafold_pytorch).
#### plmDCA
Below is a basic list of instructions for installing and preparing `plmDCA`.
It is a mixture of shell (command line) and MATLAB instructions.
It will likely need to be modified for your purposes; for instance, I downloaded the tool's directory to a different directory to keep it as a module separate from this project.
```bash
# (command line)
git clone https://github.com/magnusekeberg/plmDCA.git
cp plmDCA.m plmDCA/plmDCA_asymmetric_v2/
cd plmDCA/plmDCA_asymmetric_v2/functions/
```
```matlab
# (MATLAB)
mex calc_inverse_weights.c
mex g_rC.c
```
```bash
# (command line)
cd ../3rd_party_code/minFunc/
```
```matlab
# (matlab)
mex lbfgsAddC.c
mex lbfgsC.c
mex lbfgsProdC.c
mex mcholC.c
```
---
[](https://www.python.org)
[](https://jupyterlab.readthedocs.io/en/stable)
[](https://www.tensorflow.org)
[](https://github.com/pre-commit/pre-commit)
[](https://github.com/psf/black)