How mjlab-columbia/spade-ptm works
Project Type & Purpose
• This is a Python command-line tool built specifically to analyze SEC-MS data and evaluate protein assembly dynamics through the lens of post-translational modifications (PTMs).
• Its core functionality is implemented as a sequential pipeline with four main operations: data preprocessing, peak picking, peak matching/alignment, and post alignment editing/calculations.
• The project targets the analysis of TMT intensity tables from Spectromine, performing operations like smoothing, peak detection, alignment, and additional analytical computations (e.g., fold changes, molecular weight estimation).
File Structure Analysis
• The repository contains the following key Python scripts:
• dataPreprocessing.py: Handles initial data reshaping, normalization, averaging, and exports the results into an SQL database.
• peakPicking.py: Processes the SQL database output by smoothing the data (using SciPy’s digital filter), performing peak picking (via scipy.signal.find_peaks), and plotting peak multiplicity.
• PeakMatching.py: Aligns the peaks between samples—including matching PTM peptides to global peaks—based on a provided threshold.
• postMatchingEditing.py: Conducts post-alignment calculations such as computing mean elution fractions, fold change calculations, molecular weight estimations, and categorization of peak regions.
• Supporting files include configuration descriptors (environment.yaml, requirements.txt) and a LICENSE and README.
• Branch out from the SQL database to the “Peak Picking” module (annotate with “smoothing” and “find_peaks” operations).
• Resulting data to the “Peak Matching/Alignment” module, emphasizing configuration (e.g., threshold value).