Learn R Programming

⚠️There's a newer version (1.4.3) of this package.Take me there.

SPUTNIK (version 1.4)

SPatially aUTomatic deNoising for Ims toolKit

Description

A set of tools for the peak filtering of mass spectrometry imaging data (MSI or IMS) based on spatial distribution of signal. Given a region-of-interest (ROI), representing the spatial region where the informative signal is expected to be localized, a series of filters determine which peak signals are characterized by an implausible spatial distribution. The filters reduce the dataset dimensionality and increase its information vs noise ratio, improving the quality of the unsupervised analysis results, reducing data dimensionality and simplifying the chemical interpretation.

Copy Link

Version

Install

install.packages('SPUTNIK')

Monthly Downloads

339

Version

1.4

License

GPL (>= 3)

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/paoloinglese/SPUTNIK

Maintainer

Paolo Inglese

Last Published

October 18th, 2021

Functions in SPUTNIK (1.4)

binKmeans,msi.dataset-method

Return a binary mask generated applying k-means clustering on first 10 principal components of peaks intensities.

binSupervised,msi.dataset-method

Return a binary mask generated applying a supervised classifier on peaks intensities using manually selected regions corresponding to off-sample and sample-related areas.

binOtsu,ms.image-method

Binarize MS image using Otsu's thresholding.

closeImage,ms.image-method

Apply morphological closing to binary image.

countPixelsFilter

Filter based on the minimum number of connected pixels in the ROI.

refImageBinaryKmeansMulti

Calculate the binary reference image using k-means clustering with multi-cluster merging. K-means is run on the first `npcs` principal components to speed up the calculations.

bladderMALDIRompp2010

Load the example MALDI-MSI data.

getMZ,msi.dataset-method

Return the m/z vector.

getShapeMSI,msi.dataset-method

Returns the geometrical shape of MSI dataset

Constructor for ms.image-class objects.

refImageBinaryOtsu

Calculate the binary reference image using Otsu's thresholding.

ms.image-class definition.

invertImage,ms.image-method

Invert the colors of an MS image.

msi.dataset-class

msi.dataset-class S4 class definition containing the information about the mass spectrometry imaging dataset.

refImageContinuous

refImageContinuous returns the reference image, calculated using the refMethod. This images represents the basic measure for the filters in SPUTNIK.

plot,ms.image,missing-method

Visualize an MS image. plot extends the generic function to ms.image-class objects.

refImageBinarySVM

Calculate the binary reference image using linear SVM trained on manually selected pixels.

refImageBinaryKmeans

Calculate the binary reference image using k-means clustering. K-Means is run on the first `npcs` principal components to speed up the calculations.

smoothImage,ms.image-method

Apply Gaussian smoothing to an MS image.

Spatial chaos measure.

totalIonCountMSI,msi.dataset-method

Generates an msImage representing pixels total-ion-counts. This image can be used to qualitatively evaluate the spatial heterogeneity of the sample.

splitPeaksFilter

Test for the presence of split peaks.

globalPeaksFilter

Reference similarity based peak selection.

Constructor for msi.dataset-class objects.

createPeaksFilter

Generate a peak filter object.

getIntensityMat,msi.dataset-method

Return the peaks intensity matrix.

normIntensity,msi.dataset-method

Normalize the peaks intensities.

varTransform,msi.dataset-method

Variance stabilizing transformation.

ovarianDESIDoria2016

Load the example DESI-MSI data.

removeSmallObjects,ms.image-method

Remove binary ROI objects smaller than user-defined number of pixels

numDetectedMSI,msi.dataset-method

Generates an msImage representing the number of detected peaks per pixel. This image can be used to qualitatively evaluate the spatial heterogeneity of the sample.

Pixel scatteredness ratio.

applyPeaksFilter,msi.dataset-method

Apply the results of a peaks filter.

Performs the peak selection based on complete spatial randomness test.

PCAImage,msi.dataset-method

Generates an RGB msImage representing the first 3 principal components. This image can be used to qualitatively evaluate the spatial heterogeneity of the sample.

binKmeans2,msi.dataset-method

Return a binary mask generated applying k-means clustering on peaks intensities. A finer segmentation is obtained by using a larger number of clusters than 2. The off-sample clusters are merged looking at the most frequent labels in the image corners. The lookup areas are defined by the kernel size.

Normalized mutual information (NMI).

Structural similarity index (SSIM).