bspline_mutual_information
Utility to bin continuous variables and estimate mutual information based on B-Spline binning.
This is an adaption of Carsten Daub's R implementation1 of the algorithm described in Daub et.al 20042.
Dependencies:
- python >= 3.9
- numpy >= 1.22.0
- scipy >= 1.9.0
Installation
Recommended
The easiest way to install the package is directly via pip:
pip install bspline-mutual-information
Alternative
The most up-to-date version of the package can be also be installed by specifying the git repository directly in the pip install command
pip install git+https://github.com/pnnl-predictive-phenomics/bspline_mutual_information.git
or alternatively, cloning the repository to a local folder and installed via pip from source:
git clone https://github.com/pnnl-predictive-phenomics/bspline_mutual_information.git
cd bspline_mutual_information
pip install .
Usage
Once installed the module consits of two functions:
bspline_mutual_information.bspline_bin()bspline_mutual_information.mutual_information()
bspline_bin() is both called internally in mutual_information() but can also be used to manually bin continuous data into discrete bins if so desired.
For this example we will focus on mutual_information() since it is the primary use case of this package.
An example can be found below on how to use mutual_information() to estimate the mutual information between two vectors containing continuous data points. It assumes a python environment.
>>> from bspline_mutual_information import mutual_information
>>> x = [1,2,3,4,5]
>>> y = [1,2,1,2,3]
>>> mutual_information(x, y, bins=5, spline_order=3)
0.4740122135541802
If mutual information for pairs of columns in a whole matrix or pandas DataFrame should be calculated this can be done by leveraging pandas DataFrame.corr() function. However, since corr() does not allow for keyword arguments a helper function needs to be defined first that specifies the arguments mutual_information() should be executed with. Consider the example below where the helper function mut_inf(x, y) is defined as returning the mutual information calculated using the parameters bins=5, spline_order=1 and correct=True (which corrects for the finit size effect if spline_order==1). Note that DataFrame.corr() will always fill the diagonal with 1 values independent of the chosen method.
>>> import pandas as pd
>>> import bspline_mutual_information as bsp
>>> data = pd.DataFrame({
... 'a': [0, 1, 2, 3, 4],
... 'b': [5, 6, 7, 8, 9],
... 'c': [0, 1, 2, 1, 0],
... 'd': [3, 2, 1, 3, 2]
... })
...
>>> def mut_inf(x, y):
... mi = bsp.mutual_information(
... x, y,
... bins=5,
... spline_order=1,
... correct=True
... )
... return mi
...
>>> data.corr(method=mut_inf)
a b c d
a 1.000000 1.921928 1.121928 1.121928
b 1.921928 1.000000 1.121928 1.121928
c 1.121928 1.121928 1.000000 0.321928
d 1.121928 1.121928 0.321928 1.000000
References
-
Daub CO, Steuer R, Selbig J, Kloska S. Estimating mutual information using B-spline functions--an improved similarity measure for analysing gene expression data. BMC Bioinformatics. 2004 Aug 31;5:118. doi: 10.1186/1471-2105-5-118. PMID: 15339346; PMCID: PMC516800. ↩
Metadata
Release files for bspline-mutual-information 1.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| bspline_mutual_information-1.0.0.tar.gz | 9.2 kB | Without provenance | Details for bspline_mutual_information-1.0.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| bspline_mutual_information-1.0.0-py3-none-any.whl | Python 3 | none | any | Details for bspline_mutual_information-1.0.0-py3-none-any.whl |
Total release size: 18.9 kB
Release files / bspline_mutual_information-1.0.0.tar.gz
| Download URL | bspline_mutual_information-1.0.0.tar.gz |
|---|---|
| Size | 9.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e9a1420d06bf04f0b451fbbd98e422e8173a8e39c500122cc281856da9684222
|
|
BLAKE2b-256 checksum How to use checksums |
48602e8a882fc947062fbea17e6bde6f0b42b71080353fe3f04cdddf6ae7c127
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
python-httpx/0.27.2
|
Release files / bspline_mutual_information-1.0.0-py3-none-any.whl
| Download URL | bspline_mutual_information-1.0.0-py3-none-any.whl |
|---|---|
| Size | 9.7 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
53f59468f9fc3636ef27bd2d89033a5a60786a71d00867c3d48c7c2d718b35f6
|
|
BLAKE2b-256 checksum How to use checksums |
0cc32ab4fe678454d4dcc5f15949759a75083c99cad774322da1a93b250f5e0e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
python-httpx/0.27.2
|
Security for bspline-mutual-information 1.0.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.