This repository contains the source code for MAPPER 2.0, that was used for quantitative analysis of Drosophila wings for the manuscript "A machine-learning enhanced data fusion pipeline identifies new G-protein-based regulators of Drosophila wing growth and morphogenesis." MAPPER 2.0 provides a graphical user interface (GUI) for segmentation, feature extraction, and batch processing of adult Drosophila wings. Additional details on the workflow and GUI usage are provided in the documentation.
- Transitioning from MATLAB to a standalone Python-based GUI
- Deep learning integration (ResNet-18) for automated qualitative classification of severe phenotypes (crumpled, melanotic, domain, etc.) alongside U-Net segmentation and KNN compartment labeling
- Automated wing auto-cropping from slide-scanner outputs
- Improved quantitative morphometrics, including intervein compartment measurements, biological axes (AP/PD), crossvein defects, and cell packing/trichome density quantification
- End-to-end batch processing capabilities exporting directly to standardized CSV summary files
Before installing MAPPER 2.0, please install Anaconda on your computer from the official Anaconda website.
-
Open Anaconda Prompt.
-
Navigate to the folder where you downloaded the MAPPER 2.0 code using the
cdcommand. For example:cd Desktop cd mapper-appYou can use the following command to view the contents of your current directory:
dir -
Make sure you are inside the root code directory, where the configuration file
mapper_env.ymlis located. -
Create a new conda environment by running:
conda env create -f mapper_env.yml -n mapper_2_envThis step may take several minutes because the required packages will be downloaded and installed.
-
Once installation is complete, activate the environment:
conda activate mapper_2_env -
Run MAPPER 2.0 by executing:
python pyMAPPER-v1.py -
The MAPPER 2.0 graphical user interface (GUI) will launch automatically.