invvar

The data. A m by n matrix, where m is the number of samples and n is the number of features.

The negative controls. A logical vector of length n.

A m by (p + q) matrix containing both the factor(s) of interest (X) and known covariates (Z).

Gene-wise (as opposed to sample-wise) covariates. These covariates are adjusted for by RUV-1 before any further analysis proceeds. A matrix with n columns.

Ridge parameter. If specified, the ridged inverse method will be used.

lambda

Can be included to speed up execution. Generally used when calling invvar many times with different values of lambda.

invsvd

Estimate the features' variances using the inverse method. This function is usually called from <code>RUVinv</code> and not normally intended for stand-alone use.

Implements the 'RUV' (Remove Unwanted Variation) algorithms. These algorithms attempt to adjust for systematic errors of unknown origin in high-dimensional data. The algorithms were originally developed for use with genomic data, especially microarray data, but may be useful with other types of high-dimensional data as well. These algorithms were proposed in Gagnon-Bartsch and Speed (2012) <doi:10.1093/nar/gkz433>, Gagnon-Bartsch, Jacob and Speed (2013), and Molania, et. al. (2019) <doi:10.1093/nar/gkz433>. The algorithms require the user to specify a set of negative control variables, as described in the references. The algorithms included in this package are 'RUV-2', 'RUV-4', 'RUV-inv', 'RUV-rinv', 'RUV-I', and RUV-III', along with various supporting algorithms.

Johann Gagnon-Bartsch

Detect and Remove Unwanted Variation using Negative Controls

invvar function

<dl> <dt>Y</dt>
<dd>The data. A m by n matrix, where m is the number of samples and n is the number of features.</dd> <dt>ctl</dt>
<dd>The negative controls. A logical vector of length n.</dd> <dt>XZ</dt>
<dd>A m by (p + q) matrix containing both the factor(s) of interest (X) and known covariates (Z).</dd> <dt>eta</dt>
<dd>Gene-wise (as opposed to sample-wise) covariates. These covariates are adjusted for by RUV-1 before any further analysis proceeds. A matrix with n columns.</dd> <dt>lambda</dt>
<dd>Ridge parameter. If specified, the ridged inverse method will be used.</dd> <dt>invsvd</dt>
<dd>Can be included to speed up execution. Generally used when calling invvar many times with different values of lambda.</dd></dl>

Arguments

Johann Gagnon-Bartsch <a href="/link/johanngb%40umich.edu?package=ruv&version=0.9.7.1" data-mini-rdoc="ruv::johanngb@umich.edu">johanngb@umich.edu</a>

Author

Inverse Method Variances — invvar

<dl>

 <dt>Y</dt>
<dd>The data. A m by n matrix, where m is the number of samples and n is the number of features.</dd>

 <dt>ctl</dt>
<dd>The negative controls. A logical vector of length n.</dd>

 <dt>XZ</dt>
<dd>A m by (p + q) matrix containing both the factor(s) of interest (X) and known covariates (Z).</dd>

 <dt>eta</dt>
<dd>Gene-wise (as opposed to sample-wise) covariates. These covariates are adjusted for by RUV-1 before any further analysis proceeds. A matrix with n columns.</dd>

 <dt>lambda</dt>
<dd>Ridge parameter. If specified, the ridged inverse method will be used.</dd>

 <dt>invsvd</dt>
<dd>Can be included to speed up execution. Generally used when calling invvar many times with different values of lambda.</dd>

</dl>

Johann Gagnon-Bartsch <a href='mailto:johanngb@umich.edu'>johanngb@umich.edu</a>

invvar: Inverse Method Variances

Description

Usage

Value

Arguments

Author

References

See Also