PymolFold

Fold your protein in PyMOL!
Inspired by ColabFold by Sergey O.
Visualization inspired by pymol-color-alphafold.
Thanks to ESMFold by Meta and the API.

Tested under macOS Monterey Version 12.5.1, Python 3.7.12.
Open an issue if ran into any errors.

15Nov2022: We now provide an unofficial API to support user defined recycle number and allow sequence length up to 500aa!

Install pymol-open-source

conda install -c conda-forge pymol-open-source

Usage

1. Load extension into PyMOL. In the PyMOL command prompt:

run https://raw.githubusercontent.com/JinyuanSun/PymolFold/main/predict_structure.py
# for user still using python2, it is also py3 compatible.
run https://raw.githubusercontent.com/JinyuanSun/PymolFold/py27/predict_structure.py
# try the command below in China mainland, the mirror will be delayed if modifications were just made, download the file to your computer and install it is always a good idea:
run https://raw.staticdn.net/JinyuanSun/PymolFold/main/predict_structure.py

2. Fold your protein!

The color_plddt command also returns pymol selection object of different confidence levels. The color scheme is now compatible with plddt in range (0, 1) and (0, 100) only if they are consistent in your selection.

The Meta API (up to 400 aa):

esmfold GENGEIPLEIRATTGAEVDTRAVTAVEMTEGTLGIFRLPEEDYTALENFRYNRVAGENWKPASTVIYVGGTYARLCAYAPYNSVEFKNSSLKTEAGLTMQTYAAEKDMRFAVSGGDEVWKKTPTANFELKRAYARLVLSVVRDATYPNTCKITKAKIEAFTGNIITANTVDISTGTEGSGTQTPQYIHTVTTGLKDGFAIGLPQQTFSGGVVLTLTVDGMEYSVTIPANKLSTFVRGTKYIVSLAVKGGKLTLMSDKILIDKDWAEVQTGTGGSGDDYDTSFN, test
color_plddt
orient 
ray 1280, 960, async=1

The PymolFold API (up to 500 aa, number of recycle can be set in range (3,24)):

pymolfold GENGEIPLEIRATTGAEVDTRAVTAVEMTEGTLGIFRLPEEDYTALENFRYNRVAGENWKPASTVIYVGGTYARLCAYAPYNSVEFKNSSLKTEAGLTMQTYAAEKDMRFAVSGGDEVWKKTPTANFELKRAYARLVLSVVRDATYPNTCKITKAKIEAFTGNIITANTVDISTGTEGSGTQTPQYIHTVTTGLKDGFAIGLPQQTFSGGVVLTLTVDGMEYSVTIPANKLSTFVRGTKYIVSLAVKGGKLTLMSDKILIDKDWAEVQTGTGGSGDDYDTSFN, 4, test
color_plddt
orient 
ray 1280, 960, async=1

Reference

@article{lin2022language,
  title={Language models of protein sequences at the scale of evolution enable accurate structure prediction},
  author={Lin, Zeming and Akin, Halil and Rao, Roshan and Hie, Brian and Zhu, Zhongkai and Lu, Wenting and dos Santos Costa, Allan and Fazel-Zarandi, Maryam and Sercu, Tom and Candido, Sal and others},
  journal={bioRxiv},
  year={2022},
  publisher={Cold Spring Harbor Laboratory}
}