PathwaysModule

Delineate differential pathway modules with high biological interpretability

Biologically Explainable Machine Learning Framework for Phenotype Prediction using omics data described in Chen and Schwarz (2017) <doi:10.48550/arXiv.1712.00336>.Identifying reproducible and interpretable biological patterns from high-dimensional omics data is a critical factor in understanding the risk mechanism of complex disease. As such, explainable machine learning can offer biological insight in addition to personalized risk scoring.In this process, a feature space of biological pathways will be generated, and the feature space can also be subsequently analyzed using WGCNA (Described in Horvath and Zhang (2005) <doi:10.2202/1544-6115.1128> and Langfelder and Horvath (2008) <doi:10.1186/1471-2105-9-559> ) methods.

Shunjie Zhang

BioM2

Biologically Explainable Machine Learning Framework

Junfang Chen

PathwaysModule function

<dl><dt>pathways_matrix</dt>
<dd>A pathway matrix generated by the BioM2( target='pathways') function.</dd>
<dt>control_label</dt>
<dd>The label of the control group ( A single number, factor, or character )</dd>
<dt>power</dt>
<dd>soft-thresholding power for network construction. Detail for WGCNA::blockwiseModules()</dd>
<dt>minModuleSize</dt>
<dd>minimum module size for module detection. Detail for WGCNA::blockwiseModules()</dd>
<dt>mergeCutHeight</dt>
<dd>dendrogram cut height for module merging. Detail for WGCNA::blockwiseModules()</dd>
<dt>cutoff</dt>
<dd>Thresholds for Biological Interpretability Difference Modules</dd>
<dt>MinNumPathways</dt>
<dd>Minimum number of pathways included in the biologically interpretable difference module</dd>
<dt>p.adjust.method</dt>
<dd>p-value adjustment method.(holm", "hochberg", "hommel", "bonferroni", "BH", "BY",</dd>
<dt>exact</dt>
<dd>Whether to divide GO pathways more accurately (work when ancestor_anno=NULL)</dd>
<dt>ancestor_anno</dt>
<dd>Annotations for ancestral relationships (like data('GO_Ancestor') )</dd></dl>

Arguments

Delineate differential pathway modules with high biological interpretability — PathwaysModule

<dl>

<dt>pathways_matrix</dt>
<dd>A pathway matrix generated by the BioM2( target='pathways') function.</dd>


<dt>control_label</dt>
<dd>The label of the control group ( A single number, factor, or character )</dd>


<dt>power</dt>
<dd>soft-thresholding power for network construction. Detail for WGCNA::blockwiseModules()</dd>


<dt>minModuleSize</dt>
<dd>minimum module size for module detection. Detail for WGCNA::blockwiseModules()</dd>


<dt>mergeCutHeight</dt>
<dd>dendrogram cut height for module merging. Detail for WGCNA::blockwiseModules()</dd>


<dt>cutoff</dt>
<dd>Thresholds for Biological Interpretability Difference Modules</dd>


<dt>MinNumPathways</dt>
<dd>Minimum number of pathways included in the biologically interpretable difference module</dd>


<dt>p.adjust.method</dt>
<dd>p-value adjustment method.(holm", "hochberg", "hommel", "bonferroni", "BH", "BY",</dd>


<dt>exact</dt>
<dd>Whether to divide GO pathways more accurately (work when ancestor_anno=NULL)</dd>


<dt>ancestor_anno</dt>
<dd>Annotations for ancestral relationships (like data('GO_Ancestor') )</dd>

</dl>

PathwaysModule: Delineate differential pathway modules with high biological interpretability

Description

Usage

Value

Arguments