Maximum Entropy (MEM) re-weighting of conformational ensembles directed by the inter-dye distance distributions derived from FRET experiments.
cd example_Lif
../maxent.py \
--exp 'ucfret_20201210/*.txt' \
--weights MD/cluster_weights_ff99sb-disp.dat \
--rmps MD/Rmp_ff99sb-disp.dat \
--out result_ff99sb-disp \
--theta 0.31
First argument (--exp) is the path pattern for experimental data files. --weigths is the path to the file with reference conformer weights. --rmps is the path to the data file, which contains distance between mean positions (Rmp) for all conformers. --out is the path to the directory to save the results.
Last argument (0.31) is θ. θ is the tuning parameter that allows to chose the relative weight of χ2 and entropy (S). For θ = 0 the entropy is ignored and the algorithm minimizes the χ2, for θ = +∞, χ2 is ignored and no reweighing can happen since the original weights are fully preserved. Description of command-line arguments is also available via ./maxent.py -h.
An example of results is available at example_Lif/result_ff99sb-disp directory.
The results include the optimized cluster weights, optimized inter-dye distance distributions, and corresponding plots for each FRET pair.
More information about the Maximum Entropy Method in the context of conformational ensembles can be found in the following publication:
Köfinger J, Różycki B, Hummer G. Inferring structural ensembles of flexible and dynamic macromolecules using Bayesian, maximum entropy, and minimal-ensemble refinement methods. InBiomolecular Simulations 2019 (pp. 341-352). Humana, New York, NY.