Welcome to PEGASUS webserver

Cluster workload 0%
API workload

MD-derived protein flexibility prediction


PEGASUS (ProtEin lanGuAge models for prediction of SimUlated dynamicS) is a deep learning-based tool designed to predict protein flexibility derived from molecular dynamics simulations directly from sequence data. It utilises advanced protein language models to generate continuous predictions for key flexibility metrics, including RMSF, Std. of Phi and Psi angles, and Mean LDDT. The tool processes up to 100 sequences simultaneously of 1000 residues maximum, combining the predictions from four different PLM embeddings to enhance accuracy and reliability.

Please use the following reference when citing PEGASUS:

Yann Vander Meersche, Gabriel Duval, Gabriel Cretin, Aria Gheeraert, Jean-Christophe Gelly, Tatiana Galochkina (2025).
PEGASUS: Prediction of MD-derived protein flexibility from sequence.
Protein Science https://doi.org/10.1002/pro.70221

To launch more or very large batches of predictions please use the standalone program (Docker container) freely available on github.
Input sequences
Paste sequence example]:

Limit per job: up to 100 sequences (1000 residues max each).

Or, upload a (multi)fasta file:
Job options
Webserver Statistics
Total number of jobs submitted (since nov-2024):
Number of jobs submitted last 7 days:
Number of unique submitters: