Lightning

Accompanying code for the paper:

• Gabitto, Marie-Nelly, Pakman, Pataki, Darzak, Jordan A Bayesian nonparametric approach to super-resolution single-molecule localization

Installation

Lightning is implemented in Python 3. To install it along with the C++ code used for Quadtree acceleration, run the following:

git clone https://github.com/marianogabitto/Lightning.git
cd LightningF/Utils/QT/python/
python setup.py build_ext --inplace

For more information on the installation of the C++ code see LightningF/Utils/QT/README.txt.

Requirements

The main packages required before running the software are:

scipy, numpy, pandas, seaborn, pystan, matplotlib, scikit-learn, cython, disutils, multiprocessing

For anaconda users, the files environment.yml and requirements.txt in the root directory can be used to recreate the environment.

Usage

To use Lightning, import in your code the corresponding class:

import sys
sys.path.append('/path/to/root/of/the/Lightning/repository/')
from LightningF.Models.pdp_simple import TimeIndepentModelPython as ModelSpacePy

The input data should be created as a 4-column / 6-column numpy array. The first 2 / 3 columns correspond to x,y / x,y,z observation positions. The next 2 / 3 columns correspond to the position uncertainties s_x, s_y / s_x, s_y, s_z.

The algorithm is then run by including:

modelpy = ModelSpacePy(data=data, init_type='rl_cluster', infer_pi1=True, infer_alpha0=True)
modelpy.fit(iterations=100, pl=0, prt=1)

And the results can be visualized with a plot generated with:

modelpy.pl_bl() 

Example

An example showing how to run Lightning appears in the file comparison.py, located in the directory MCMC_compare. To run it, execute:

cd MCMC_compare
python comparison.py

The code loads one of the datasets used in the paper, compares Lightning with MCMC, prints several metrics and creates a pdf file in the same directory entitled inference_50nm.pdf, corresponding to Supplementary Figure 2 from the paper.

Citation

If you use this project, please cite the relevant publication as:

@article{gabitto2020bayesian,
  title={A Bayesian nonparametric approach to super-resolution single-molecule localization},
  author={Gabitto, Mariano Ignacio and Marie-Nelly, Herve and Pakman, Ari and Pataki, Andras and Darzacq, Xavier and Jordan, Michael},
  journal={bioRxiv},
  year={2020},
  publisher={Cold Spring Harbor Laboratory}
}