PLMSearch

2024.6.5 Update: We have uploaded the Dataset of PLMSearch & PLMAlign in Zenodo.
2024.5.30 Update: We have uploaded the Dataset of PLMSearch Web Server in Zenodo.

This is the implement of "PLMSearch: Protein language model powers accurate and fast sequence search for remote homology". By using a protein language model, PLMSearch can achieve a sensitivity close to SOAT structure search methods while being versatile and fast because it is only based on sequences.

Quick links

Webserver
Requirements
Data preparation
Reproduce all our experiments with only one file
Run PLMSearch locally
Citation

Webserver

PLMSearch web server : dmiip.sjtu.edu.cn/PLMSearch 🚀

PLMAlign web server : dmiip.sjtu.edu.cn/PLMAlign :airplane:

PLMAlign source code : github.com/maovshao/PLMAlign :helicopter:

Requirements

Follow the steps in requirements.sh

Data preparation

We have released our experiment data, which can be downloaded from plmsearch_data or Zenodo.

# Include experiment data, PLMSearch model, ESM-1b model, etc.
# Use the following command or download it from https://zenodo.org/records/11480660
wget https://dmiip.sjtu.edu.cn/PLMSearch/static/download/plmsearch_data.tar.gz  
tar zxvf plmsearch_data.tar.gz

Reproduce all our experiments with only one file

Reproduce all our experiments with good visualization by following the steps in main.ipynb

Notice: Detailed results are saved in scientist_figures/.

Run PLMSearch locally

Run PLMSearch locally by following the example in pipeline.ipynb

Notice: the inputs and outputs of the example are saved in example/.

Citation

Liu, W., Wang, Z., You, R. et al. PLMSearch: Protein language model powers accurate and fast sequence search for remote homology. Nat Commun 15, 2775 (2024). https://doi.org/10.1038/s41467-024-46808-5

maovshao / PLMSearch

readme