waltlabtools

A collection of tools for biomedical research assay analysis in Python.

Key Features

• Analysis for assays such as digital ELISA.

• Calculation of calibration curves, concentrations, limits of detection, and more.

• Free and open-source software under the GNU General Public License v3.

Getting Started

• Installation: waltlabtools can be installed using conda or pip. In the command line,

  • conda: conda install -c tylerdougan waltlabtools

  • pip: pip install waltlabtools

• Dependencies: waltlabtools requires scipy ≥ 1.3.1, and either jax ≥ 0.1.64 or numpy ≥ 1.10.0. To make the best use of waltlabtools, you may want to install pandas (for data import/export and organization), scikit-learn (for data analysis), matplotlib (for plotting), and JupyterLab (for writing code). These can all be installed using conda or pip, and may become dependencies in future releases.

Functions and Classes

API: waltlabtools includes classes for mathematical models and calibration curves, and a set of functions to make use of these objects and others. These are covered in the documentation.

noindex

Functions

aeb(fon)	The average number of enzymes per bead.
c4(n)	Factor c4 for unbiased estimation of the standard deviation.
flatten(data[, on_bad_data])	Flattens most data structures.
fon(aeb)	The fraction of beads which are on.
gmnd(data)	Geometric meandian.
jit(fun)
lod(blank_signal[, inverse_fun, sds, corr])	Compute the limit of detection (LOD).
regress(model, x, y[, use_inverse, weights, ...])	Performs a (nonlinear) regression and return coefficients.

Classes

CalCurve([model, coefs, lod, lod_sds, force_lod])	Calibration curve.
Model([fun, inverse, name, params, xscale, ...])	Mathematical model for calibration curve fitting.

noindex

Functions

Id(x)

The identity function.

---

Development of waltlabtools is led by the Walt Lab for Advanced Diagnostics at Brigham and Women's Hospital, Harvard Medical School, and the Wyss Institute for Biologically Inspired Engineering.