rapids_singlecell.dcg.mlm

Contents

rapids_singlecell.dcg.mlm#

rapids_singlecell.dcg.mlm = <rapids_singlecell.decoupler_gpu._helper._Method.Method object>[source]#

Multivariate Linear Model (MLM).

This approach uses the molecular features from one observation as the population of samples and it fits a linear model with with multiple covariates, which are the weights of all feature sets \(F\).

\[y^i = \beta_0 + \beta_1 x_{1}^{i} + \beta_2 x_{2}^{i} + \cdots + \beta_p x_{p}^{i} + \varepsilon\]

Where:

  • \(y^i\) is the observed feature statistic (e.g. gene expression, \(log_{2}FC\), etc.) for feature \(i\)

  • \(x_{p}^{i}\) is the weight of feature \(i\) in feature set \(F_p\). For unweighted sets, membership in the set is indicated by 1, and non-membership by 0.

  • \(\beta_0\) is the intercept

  • \(\beta_p\) is the slope coefficient for feature set \(F_p\)

  • \(\varepsilon\) is the error term for feature \(i\)

The enrichment score \(ES\) for each \(F\) is then calculated as the t-value of the slope coefficients.

\[ES = t_{\beta_1} = \frac{\hat{\beta}_1}{\mathrm{SE}(\hat{\beta}_1)}\]

Where:

  • \(t_{\beta_1}\) is the t-value of the slope

  • \(\mathrm{SE}(\hat{\beta}_1)\) is the standard error of the slope

Next, \(p_{value}\) are obtained by evaluating the two-sided survival function (\(sf\)) of the Student’s t-distribution.

\[p_{value} = 2 \times \mathrm{sf}(|ES|, \text{df})\]
Parameters:
data

AnnData instance, DataFrame or tuple of [matrix, samples, features].

net

Dataframe in long format. Must include source and target columns, and optionally a weight column.

tmin default: 5

Minimum number of targets per source. Sources with fewer targets will be removed.

layer

Layer key name of an anndata.AnnData instance.

raw default: False

Whether to use the .raw attribute of anndata.AnnData.

empty default: True

Whether to remove empty observations (rows) or features (columns).

bsize default: 5000

For large datasets in sparse format, this parameter controls how many observations are processed at once. Increasing this value speeds up computation but uses more memory.

verbose default: False

Whether to display progress messages and additional execution details.

pre_load default: False

Whether to pre-load the data into memory. If True, the data will be pre-loaded into memory before processing.

adj_pv_gpu default: False

Whether to use GPU for adjusting p-values.

tval

Whether to return the t-value (tval=True) the coefficient of the fitted model (tval=False).

Returns:

Enrichment scores \(ES\) and, if applicable, adjusted \(p_{value}\) by Benjamini-Hochberg.

Example

import decoupler as dc

adata, net = dc.ds.toy()
rsc.dcg.mlm(adata, net, tmin=3)
Contents