Scripts to analyze raw H5 files of test data from Optical tweezer DNA unwinding experiment

C-trap DNA Unwinding Analysis Python scripts for analyzing DNA unwinding by a viral helicase from single-molecule optical-tweezers (C-trap) data collected on the LUMICKS platform. Raw traces are read from LUMICKS .h5 files, the trap-to-trap distance is converted into nucleotides (nt) unwound, and the resulting time traces are plotted, exported, and (optionally) analyzed for unwinding length and pausing behavior. Raw traces for testing the script are provided in the repository. What the scripts do Script Purpose Marker_Data_analysis_Ctrap_1.py Load .h5 files from a folder, convert distance → nt unwound, plot distance vs. time, and batch-export the plots as PDFs. PauseAndFinalUnwindingLegnth_forSpecificTime.py For each trace, smooth the signal, measure the total nt unwound at a fixed time point (default 60 s), and estimate the fraction of the trace spent paused. Requirements Python 3.9+ lumicks.pylake — reading .h5 files numpy pandas matplotlib scipy — Savitzky–Golay / median smoothing pyperclip — used by the pause script to read the folder path from the clipboard Install everything with: pip install lumicks.pylake numpy pandas matplotlib scipy pyperclip Note: The scripts were written to run interactively (e.g. in Spyder or a Jupyter/IPython session). Marker_Data_analysis_Ctrap_1.py uses the IPython magic %matplotlib auto and cell markers (#%%), and the pause script assumes numpy, pandas, matplotlib.pyplot, and lumicks.pylake are already imported in the session. If you run them as plain .py files, add the standard imports at the top (see below). Usage 1. Plot and export traces — Marker_Data_analysis_Ctrap_1.py Set the folder that contains your .h5 files. In the script this is the path variable used in os.chdir(path) — assign it before running the file-listing cell, e.g.: path = r"C:\path\to\your\h5\folder" Run the file-listing cell. It collects every .h5 file in the folder and prints them as a table. Inspect a single file: set k to the index of the file you want and run the selection cell to plot distance (nt) vs. time (s). Batch export: run the final cell to loop over every file in the folder and save each trace as <filename>Dist.pdf in the same folder. The y-axis is fixed to [-50, 500] nt for the batch plots — adjust axs.set_ylim([...]) if your traces fall outside that range. 2. Measure unwinding length and pausing — PauseAndFinalUnwindingLegnth_forSpecificTime.py Copy the folder path containing your .h5 files to the clipboard. The script reads it with pyperclip.paste(), so no editing is needed — just copy the path before running. Run the script. For every trace longer than 60 s it will: Smooth the signal with a Savitzky–Golay filter (window_length=51, polyorder=2). Record the nt unwound at the target time (target_t = 60 s) → appended to finalUnwindingLength. Compute the instantaneous rate (slope) and classify each point as paused when |slope| < flat_threshold (default 2.0 nt/s). Append the paused fraction (%) of the trace to finalPauseDuration and print it. Plot the smoothed trace (black) with paused points highlighted (yellow). Results accumulate across all files in the two arrays: finalUnwindingLength — nt unwound at 60 s for each qualifying trace finalPauseDuration — percentage of the trace spent paused Parameters you may want to tune: Parameter Default Meaning target_t 60 Time point (s) at which unwinding length is measured. time[-1] > 60 60 Minimum trace duration (s) required for a file to be analyzed. window_length / polyorder 51 / 2 Savitzky–Golay smoothing window and polynomial order. flat_threshold 2.0 Rate (nt/s) below which the helicase is considered paused. Suggested imports (for standalone runs) If you run the scripts outside of an interactive session where these are already loaded, add the following at the top of each file: import os import numpy as np import pandas as pd import matplotlib.pyplot as plt import lumicks.pylake as lk from scipy.signal import medfilt, savgol_filter import pyperclip # pause script only Data assumptions Input files are LUMICKS .h5 files containing a distance1 channel. Distance is read via file.distance1.data and time via file.distance1.seconds. All .h5 files in the selected folder belong to the same substrate/calibration (they share the conversion constants). Output Marker_Data_analysis_Ctrap_1.py — one PDF per trace (distance in nt vs. time) saved alongside the source .h5 files. PauseAndFinalUnwindingLegnth_forSpecificTime.py — in-session plots plus the finalUnwindingLength and finalPauseDuration arrays for downstream analysis or export.