Skip to main content

Estimate intracellular pH from fluorescent indicator data using bidirectional sigmoid curve fitting

Project description

pHfit

GitHub License GitHub Release GitHub Release Date Tests codecov Create Release Publish PyPI Python PyPI PyPI - Downloads Docker Docker Pulls Docker Image Size

Estimate intracellular pH from fluorescent indicator data using bidirectional sigmoid curve fitting.

Linear fit vs Sigmoid fit comparison

The relationship between pH and fluorescence signal (intensity or ratio) of pH-sensitive indicators is inherently sigmoidal, governed by the Henderson-Hasselbalch equation. While a linear approximation may appear adequate over a narrow pH range near the pKa, it systematically misestimates pH at both ends of the dynamic range — precisely where many biological samples fall. A sigmoid fit faithfully captures the saturation behavior at extreme pH values, yielding accurate estimates across the full range of the indicator.

Overview

pHfit fits a Henderson-Hasselbalch-derived sigmoid function to standard curve data from fluorescent pH indicator reagents (e.g., LysoSensor, Oregon Green, pHrodo) and estimates sample pH values from the fitted curve.

Sigmoid function:

$$F(\text{pH}) = y_{\min} + \frac{y_{\max} - y_{\min}}{1 + 10^{n \cdot (pK_a - \text{pH})}}$$

Parameter Description
y_min Minimum signal (fully protonated)
y_max Maximum signal (fully deprotonated)
pKa Apparent acid dissociation constant (inflection point)
n Hill coefficient (positive = ascending, negative = descending)

pHfit supports both ascending indicators (pH↑ → fluorescence↑, e.g., Oregon Green, BCECF) and descending indicators (pH↑ → fluorescence↓, e.g., pHrodo, LysoSensor). The direction is auto-detected from the data or inferred from the chosen preset.

Installation

pip install pHfit

For development:

git clone https://github.com/NaotoKubota/pHfit.git
cd pHfit
pip install -e ".[dev]"

Usage

Quick start with example data

Example input files and expected output are included in the example/ directory. Try it out:

phfit -i example/standard_curve.tsv -s example/sample.tsv -o example/output --preset oregongreen488

Basic usage

phfit -i standard_curve.tsv -s sample.tsv -o output/

With a reagent preset

phfit -i standard_curve.tsv -s sample.tsv -o output/ --preset oregongreen488

Descending indicator (e.g., pHrodo)

pHrodo and similar descending indicators produce higher signal at lower pH. pHfit auto-detects the direction, or you can use a preset:

# Auto-detection
phfit -i phrodo_standard.tsv -s phrodo_sample.tsv -o output/

# With preset
phfit -i phrodo_standard.tsv -s phrodo_sample.tsv -o output/ --preset phrodo_red

A pHrodo example is included in the repository:

phfit -i example/phrodo_standard_curve.tsv -s example/phrodo_sample.tsv -o example/phrodo_output

Standard curve fitting only (no samples)

phfit -i standard_curve.tsv -o output/

List available presets

You can check all available presets and their directions with:

phfit --preset list

Estimate Hill coefficient from data

phfit -i standard_curve.tsv -s sample.tsv -o output/ --hill fit

Input File Format

standard_curve.tsv

Tab-separated file with columns pH and value. Replicates (rows with the same pH) are automatically averaged.

The value column accepts any scalar fluorescence readout — single-wavelength intensity, fluorescence ratio, or any other derived metric. The same sigmoid model applies to all cases.

pH	value
3.0	105.2
3.0	110.8
3.0	98.5
4.0	150.3
4.0	145.7
4.0	155.1
5.0	450.2
5.0	448.9
5.0	455.3
6.0	720.1
6.0	715.8
6.0	725.4
7.0	790.5
7.0	785.2
7.0	792.8

sample.tsv

Tab-separated file with columns sample and value. Replicates (rows with the same sample name) are automatically averaged.

sample	value
sampleA	250.3
sampleA	245.7
sampleA	255.1
sampleB	520.4
sampleB	518.9
sampleB	525.3

Creating TSV files from Excel

If you are not familiar with TSV files, you can easily create them from Microsoft Excel or Google Sheets:

  1. Open Excel (or Google Sheets) and enter your data with the column headers (pH and value, or sample and value) in the first row.
  2. Excel: Go to File → Save As, choose Text (Tab delimited) (*.txt) as the file format, and save. Then rename the file extension from .txt to .tsv.
  3. Google Sheets: Go to File → Download → Tab-separated values (.tsv).

[!Note] Make sure the first row contains exactly the column headers pH and value (for the standard curve) or sample and value (for sample data). Do not include extra blank rows or columns.

Output Files

File Description
fit_params.tsv Fitted parameters (y_min, y_max, pKa, n, R²)
standard_curve.pdf Standard curve plot (PDF)
standard_curve.png Standard curve plot (PNG)
estimated_pH.tsv Estimated pH for each sample (averaged)
estimated_pH_all.tsv Estimated pH for each individual replicate
sample_estimates.pdf Sample estimates plot (PDF)
sample_estimates.png Sample estimates plot (PNG)
report.html Self-contained HTML report with interactive plots
summary.json Machine-readable run summary (version, timestamp, fit parameters, out-of-range statistics)

Interactive HTML Report

report.html is a self-contained, single-file HTML report that can be opened in any web browser. It includes:

  • Fitted parameters — y_min, y_max, pKa, n (Hill coefficient), R², and sigmoid direction
  • Interactive standard curve — Plotly-powered chart with hover tooltips showing exact pH and signal values; zoom, pan, and export are built-in
  • Sample pH estimates — Interactive chart mapping each sample's fluorescence to estimated pH on the fitted curve (when -s is provided)
  • Sigmoid equation — The fitted model equation displayed in the report header

No internet connection is required after generating the report — all Plotly assets are loaded via CDN at generation time and the chart data is embedded inline.

See an example report: example/output/report.html

CLI Options

Option Description
-v, --version Show version and exit
-i, --input Path to standard curve TSV (required)
-s, --sample Path to sample TSV (optional)
-o, --output Output directory (required)
--preset Reagent preset name or list
--pka Fix pKa value
--ymin Fix y_min value
--ymax Fix y_max value
--hill Hill coefficient: number to fix, or fit to estimate
--dpi PNG resolution (default: 300)
--verbose Enable verbose (DEBUG-level) logging output
--include-out-of-range Include out-of-range replicates in sample mean/SD calculation (default: excluded)

Available Presets

Preset name Reagent pKa Direction
oregongreen488 Oregon Green 488 4.7 ↑ ascending
oregongreen514 Oregon Green 514 4.8 ↑ ascending
phrodo_red pHrodo Red 6.5 ↓ descending
phrodo_green pHrodo Green 6.5 ↓ descending
lysosensor_green LysoSensor Green DND-189 5.2 ↓ descending
lysosensor_blue LysoSensor Blue DND-167 5.1 ↓ descending
lysosensor_yellowblue LysoSensor Yellow/Blue DND-160 4.2 ↓ descending
bcecf BCECF 6.98 ↑ ascending
fitc FITC / Fluorescein 6.4 ↑ ascending
snarf1 SNARF-1 7.5 ↑ ascending
cypher5e CypHer5E 7.3 ↓ descending
hpts HPTS (Pyranine) 7.3 ↑ ascending

Note on Ratiometric Indicators

Some pH indicators (e.g., BCECF, SNARF-1, HPTS, LysoSensor Yellow/Blue) are measured ratiometrically — the ratio of fluorescence at two excitation or emission wavelengths is used instead of a single intensity value. The Henderson-Hasselbalch sigmoid model applies equally to ratio data; users simply provide the pre-computed fluorescence ratio in the value column of the input TSV.

[!Tip] For ratiometric indicators, the preset pKa represents a typical apparent pKa that may vary depending on your optical setup (filter sets, detector gains, etc.). Consider omitting --pka or using --hill fit to let pHfit estimate all parameters freely from your calibration data.

License

MIT License

Citation

If you use pHfit in your research, please cite it. Citation metadata is provided in CITATION.cff. A Zenodo DOI for each released version is available from the GitHub Releases page.

Contributing

Thank you for wanting to improve pHfit! If you have any bugs or questions, feel free to open an issue or pull request. See CONTRIBUTING.md for development workflow and release process.

Authors

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

phfit-1.0.0.tar.gz (130.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

phfit-1.0.0-py3-none-any.whl (23.2 kB view details)

Uploaded Python 3

File details

Details for the file phfit-1.0.0.tar.gz.

File metadata

  • Download URL: phfit-1.0.0.tar.gz
  • Upload date:
  • Size: 130.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.6

File hashes

Hashes for phfit-1.0.0.tar.gz
Algorithm Hash digest
SHA256 a092e46eccaa2293b24c0b5af47230ec616646a3c654edfa110e84e69320d781
MD5 7f47eb9fc99a0ec315d79a7bfee6c52b
BLAKE2b-256 dc76b459f84761e3c28837cad7b33a10053baceb8c571ad815a02865aa85fd4f

See more details on using hashes here.

File details

Details for the file phfit-1.0.0-py3-none-any.whl.

File metadata

  • Download URL: phfit-1.0.0-py3-none-any.whl
  • Upload date:
  • Size: 23.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.6

File hashes

Hashes for phfit-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 9feb8ec7ec89b3c3f566dc5bd8e21758df12d644f8754bcfa083058674a139ba
MD5 8ce429739f2540581308ca9997387798
BLAKE2b-256 26947741043b5e8127d21811b06e15e886915d15704a789065060e9338c41e91

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page