Welcome to FRETraj#
FRETraj is a Python module for predicting FRET efficiencies by calculating multiple accessible-contact volumes (multi-ACV) to estimate donor and acceptor dye dynamics. The package features a user-friendly PyMOL plugin [1] for FRET-assisted, integrative structural modeling. It interfaces with the LabelLib library for fast computation of ACVs. Specifically, FRETraj is designed to:
plan (single-molecule) FRET experiments by optimizing labeling positions
interpret FRET-based distance measurements on a biomolecule
integrate FRET experiments with molecular dynamics simulations

Fig. 1 Schematic of FRETraj: A fluorophore (here Cy3) is parametrized by five distances (\(R_1-R_3\), \(L_\text{linker}\) and \(W_\text{linker}\)). The extent of surface stacking is assessed by fluorescence anisotropy decays. Accessible-contact volume calculations and FRET predictions are implemented in PyMOL (with a GUI) and Jupyter notebooks (for large-scale analyses).#