weightNorm

Use weightNorm to normalize the SVM weights. Used for SVMDECONV
w1 &lt;- weightNorm(w)

Tools to construct (or add to) cell-type signature matrices using flow sorted or single cell samples and deconvolve bulk gene expression data.
Useful for assessing the quality of single cell RNAseq experiments, estimating the accuracy of signature matrices, and determining cell-type spillover.
Please cite: Danziger SA et al. (2019) ADAPTS: Automated Deconvolution Augmentation of Profiles for Tissue Specific cells <doi:10.1371/journal.pone.0224693>.

Samuel Danziger

ADAPTS

Automated Deconvolution Augmentation of Profiles for Tissue
Specific Cells

weightNorm function

<dl><dt>w</dt>
<dd>The weight vector from fitting an SVM, something like something like t(fit1$coefs) %*% fit1$SV, where fit comes from &lt;- svm(m~B, nu=0.25, kernel="linear"))</dd></dl>

Arguments

SVMDECONV helper function — weightNorm

<dl>

<dt>w</dt>
<dd>The weight vector from fitting an SVM, something like something like t(fit1$coefs) %*% fit1$SV, where fit comes from &lt;- svm(m~B, nu=0.25, kernel="linear"))</dd>

</dl>

weightNorm: SVMDECONV helper function

Description

Usage

Value

Arguments