Blinking-Based Multiplexing (BBM) for Single Molecule Localised Microscopy (SMLM) Software

This repository contains a Python-based software for analyzing Single-Molecule Localization Microscopy (SMLM) data using Blinking-Based Multiplexing. The analysis includes background removal, signal normalization, particle localization, intensity extraction, and photon emission calculations. The software is optimized for Python 3.10.

Features

• Blinking-Based Multiplexing: Uses temporal blinking patterns of fluorophores to differentiate molecules.
• Background Removal: Utilizes sep.Background to remove noise and improve the signal-to-noise ratio.
• Signal Normalization: Normalizes intensity data to ensure consistency across datasets.
• Particle Localization: Identifies local maxima corresponding to single molecules using OpenCV.
• Intensity Extraction: Extracts raw camera counts to calculate photon emission rates and photobleaching dynamics.
• Photon Emission Calculations: Converts EMCCD counts to photon arrival rates using a custom equation.

File Structure

BlinkSMLM/
│
├── src/                         
│   ├── BBM_Functions.py         # Python functions for blink detection and background removal
│   ├── GUIs.py                  # GUI for the analysis (if applicable)
│   ├── Main.py                  # Main analysis script
│   ├── utils/                   # Utility files for efficiency and colormap
│       ├── fire_cmap.py         # Colormap generation script
│       ├── Objective_Efficiency.txt  # Objective efficiency values
│       ├── Quantum_Efficiency_iXon_897.txt  # Quantum efficiency for iXon camera
│
├── data/                        
│   ├── Demo_Data.zip            # Example input data (ziped ND2 movie file)
│
├── docs/                        
│   ├── README.md                # Main project documentation
│
├── requirements.txt             # Python dependencies list
├── LICENSE                      # License file

Installation

1. Clone this repository:

   git clone https://github.com/kirill-kniazev/BlinkSMLM.git
   cd BlinkSMLM

2. Install the required dependencies:

   pip install -r requirements.txt

Here's a refined and more polished version of the Usage section based on your input:

---

Usage

1. Launching the GUI:

To begin the analysis, open the graphical user interface (GUI) by running the Main.py script:

python src/Main.py

Once the GUI, titled "Analysis Workflow", is launched, you will see two options: "Start" to begin the analysis or "Close" to exit the process. To proceed with the analysis, click the Start button.

This action will initialize the analysis workflow, and a status message "Data Analysis in Process" will appear. A new window titled "Open the file" will be displayed.

2. Data Input (Drag and Drop):

In the "Open the file" window, you will be prompted to drag and drop your .nd2 or .tif file into the designated area. Once you do this, the input GUI will close, and a new window called "Parameters Input" will open.

3. User Input Configuration:

After selecting the data file, a popup window will appear, asking for specific parameters needed for the analysis. These include:

• Time interval between frames acquired (default: 11 ms)
• Number of frames to process (default: full data stack)
• Camera pre-amplification gain (default: 5.1)
• Camera EM gain (default: 285)
• Wavelength of emission for the dye in the optical range (665-705 nm, default: 677 nm for SF8(D4)₂)

The window will have two buttons: "Continue" to proceed and "Close" to exit.

Once you click Continue, the next step for thresholding will begin.

4. Thresholding for Particle Localization:

In the "Select Threshold to Localize Particles" window, you will define the threshold values for particle localization. The GUI provides a visual interface that helps you adjust the threshold to detect particles accurately. Ideally, you should slightly oversample the particles, and a good starting number is around 100 particles.

After setting the threshold, click Continue, and the analysis will begin. A "Trace Processing" window will pop up, displaying the processing progress in real-time.

5. Results:

Once the analysis is complete, the software will create a "Results" folder in the same directory as the file being analyzed.

Within this folder, there will be two groups of files:

• "False Positive": Contains particles that were detected but did not exhibit any photobleaching or a two-state process.
• "Positive": Contains particles that exhibited clear multi-optical processes.

These results will help you differentiate between valid and invalid single-molecule signals.

Example Input Data

The data/SN9_crop.tif file is an example input file. Replace this with your own .nd2 file for analysis.

Photon Calculation: Converting Counts to Photons

The analysis involves two steps to convert raw camera counts to the number of emitted photons from a sample. Here's a breakdown of each step:

1. Converting Camera Counts to Received Photons

The first step is to convert the raw counts detected by the camera into the number of photons that actually arrived at the camera chip. This is achieved using the following formula:

N_received = ((raw_count - bias_offset) * pre_amp_gain) / (EM_gain * Quantum_efficiency)

Where:

• raw_count is the raw signal from the camera in counts.
• bias_offset is the bias offset from the EMCCD, typically around 200 counts.
• pre_amp_gain is the pre-amplifier gain of the camera (default: 5.1).
• EM_gain is the electron-multiplying gain, typically set to 285.
• Quantum_efficiency is the camera's efficiency at converting incoming photons into counts, dependent on the wavelength.

2. Converting Received Photons to Emitted Photons

Once you have the number of photons received by the camera, the next step is to calculate the number of photons that were emitted by the sample. Not all emitted photons are captured by the camera, so we need to account for the light collection efficiency and the optical losses.

The formula for converting received photons to emitted photons is:

N_emitted = N_received / (η_coll * η_opt)

Where:

• η_coll is the collection efficiency of the microscope objective, calculated based on the numerical aperture (NA) and refractive index (n) of the sample medium, representing the fraction of emitted light collected by the objective.
• η_opt is the overall optical efficiency, accounting for transmission losses in the optical path, which includes the objective and other optical elements.

This two-step process allows you to estimate the number of photons emitted by your sample based on the raw counts recorded by the camera.