Skip to content

Home

Jump to bottom
JMB edited this page Nov 26, 2025 · 6 revisions

SAXSier Analysis Suite - User Manual

SAXSier is a Python-based graphical software suite designed for the processing and analysis of Small-Angle X-ray Scattering (SAXS) data. It is particularly optimized for SEC-SAXS (Size-Exclusion Chromatography coupled SAXS) and high-throughput analysis.

The idea is "To view everything at a glance"

Table of Contents

Overview

The suite consists of a central launcher and four specialized tools:

  1. SubMe: Buffer subtraction.
  2. Ragtime: SEC-SAXS frame-by-frame analysis (Guinier, Rg, MW).
  3. Sexier: Deep analysis of single profiles (Guinier, BIFT P(r), Vc MW).
  4. SAXSting: High-throughput comparison, overlay, and figure generation.

1. Getting Started

Installation & Requirements

The suite is written in Python and uses the PySide6 framework for the GUI and Matplotlib for plotting.

Dependencies:

  • Python 3.x
  • PySide6
  • matplotlib
  • numpy
  • scipy
  • numba (required for Sexier's BIFT calculation)

Launching the Suite

Run the main launcher script to access all tools from a central hub:

python SAXSier-v4.py
image

2. SubMe (v3.6)

Purpose: Background/Buffer subtraction for SAXS profiles.

Key Features

  • Auto-Unit Detection: Detects if data is in nm⁻¹ and converts to Å⁻¹.
  • Three Subtraction Methods:
    1. Average Buffer: Averages a user-defined range of buffer frames.
    2. Auto Linear Baseline: Uses the first 10 and last 10 frames to calculate a drifting baseline.
    3. Manual Linear Baseline: Allows the user to select specific start and end frames to define a baseline.

Workflow

  1. Load Data: Click Select Data Folder.
  2. Choose Method:
    • Tab 1 (Average): Enter Start Frame and End Frame. Click Subtract Average Buffer.
    • Tab 2 (Baseline):
      • Auto: Click Subtract Auto Baseline.
      • Manual: Use the arrow buttons (<<, <, >, >>) to select two specific frames (F1 and F2) that define the baseline. Click Subtract Manual Baseline.
  3. Output: Files are saved in a new folder created in the parent directory (e.g., SUB-Man-Avg-XX-YY or SUB-Auto-BL).
image image

3. Ragtime (v5.0)

Purpose: Primary analysis for SEC-SAXS chromatograms. Analyzes Radius of Gyration (Rg) and Molecular Weight (MW) across all frames.

Key Features

  • Auto-Unit Detection: Detects if data is in nm⁻¹ and converts to Å⁻¹.
  • Best File Detection: Automatically identifies the frame with the highest total integral.
  • Auto-Guinier: Automatically searches for the optimal linear Guinier region.
  • Peak Averaging: Extracts and averages frames from the elution peak with error propagation.

Workflow

  1. Load Data: Click Select Data Folder (select the folder containing subtracted .dat files).
  2. Auto-Analysis: The software automatically finds the "Best" file and performs an initial Guinier fit.
  3. Refinement:
    • Use the <- and -> buttons to navigate frames.
    • Manually adjust qmin index and qmax index if necessary.
    • Click Update Ragtime to apply changes to the current frame.
  4. Batch Processing:
    • (Optional) Set a Ragtime Analysis Range (Start/End frame) to limit analysis to the peak.
    • Click Run Ragtime to apply the current Guinier settings to all files.
  5. Results:
    • Generates a 2-panel plot: I(0) vs Frame overlaid with Rg and MW.
    • Saves Ragtime_Results.txt in the Ragtime-Results folder.
  6. Peak Extraction:
    • Enter First frame and Last frame of the peak.
    • Click Extract & Average Peak. This creates a high-quality averaged file in a Peak-X folder.
image

4. Sexier (v7.2)

Purpose: In-depth analysis of a single scattering curve (usually the averaged peak file from Ragtime).

Key Features

  • Auto-Unit Detection: Detects if data is in nm⁻¹ and converts to Å⁻¹.
  • Guinier Analysis: Calculates Rg and I(0).
  • Molecular Weight (Vc): Estimates MW using the Volume of Correlation method.
  • P(r) Analysis (BIFT): Bayesian Inverse Fourier Transform to calculate the Pair-Distance Distribution Function, Dmax, and Rg(real space).
  • Diagnostic Plots: Generates publication-ready 4-panel plots (Form Factor, Guinier, Residuals, Kratky, Vc).

Workflow

  1. Load Data: Load a single .dat file (e.g., the output from Ragtime).
  2. Guinier Region: The tool attempts to auto-detect the region. Adjust qmin and qmax indices manually and click Re-Process Data.
  3. MW Calculation:
    • Enter MW from sequence (kDa) to calculate Oligomeric State (e.g., Monomer/Dimer).
    • The Vc-based MW is displayed automatically.
  4. P(r) Analysis (BIFT):
    • Enter an estimated Dmax (Å) (or leave blank to auto-estimate from Rg).
    • (Optional) Set q_min / q_max limits.
    • Click Run BIFT.
  5. Output: All plots (PNG/SVG) and fit data (.ift, .txt) are saved automatically in a _Sexier_out folder.
image

5. SAXSting (v5.0)

Purpose: Comparison, Superimposition, and Figure Generation.

Key Features

  • Auto-Unit Detection: Detects if data is in nm⁻¹ and converts to Å⁻¹.
  • Drag & Drop: Easily load multiple .dat files to compare them.
  • Superimpose: One-click normalization by I(0) to overlay curves.
  • Log-Log Toggle: Quickly switch visualization modes.
  • Batch Guinier: Can run the "Sexier" auto-algorithm on a list of files.
  • Averaging: Can average a selection of normalized files.

Workflow

  1. Load Data: Drag files onto the list or use the browse button.
  2. Analyze:
    • Click Auto-Process to calculate Rg and I(0) for all loaded files.
    • Or refine specific files manually.
  3. Compare:
    • Click Superimpose to normalize curves.
    • Toggle Log-Log Plot to inspect features.
  4. Export:
    • Click Save Full Report.
    • This generates a folder containing:
      • Comparison Curves (Plots).
      • Summary Table (Excel-compatible text).
      • Source Data (Individual text files for replotting in Origin/Excel).
image

Technical Details & Mathematics

Guinier Approximation: $$ ln(I(q)) = ln(I(0)) - \frac{R_g^2}{3} \cdot q^2 $$

Molecular Weight (Vc): Calculated using the Volume of Correlation ($V_c$) and Porod Invariant ($Q_R$). $$ V_c = \frac{I(0)}{\int q \cdot I(q) dq} $$ $$ MW \approx \frac{Q_R}{0.1231} $$ (0.1231 is the empirical constant for proteins).

BIFT (Bayesian Inverse Fourier Transform): Uses a Bayesian approach to infer the P(r) distribution, minimizing the need for manual alpha-parameter tuning compared to traditional methods (like GNOM).

Clone this wiki locally