CAM3.0

CAM3.0 performs fully unsupervised deconvolution of bulk tissue data to estimate cell type compositions and cell-specific expression profiles.

Key Features:

• Radius-fixed clustering: Identifies reliable cell markers from bulk tissue data.
• Linear programming for initial scatter simplex detection: Detects an initial scatter simplex to support deconvolution.
• Smart floating search: Searches for the optimum latent variable model.
• Cell marker identification: Accurately identifies known or novel cell markers.
• Cell proportion determination: Determines cell type proportions in complex tissues.
• Cell-specific expression estimation: Estimates expression profiles specific to individual cell types.

Scientific Applications:

• Bulk tissue deconvolution without references: Deconvolves bulk tissue data when study- or datatype-specific references are unreliable or unavailable.
• Cell marker discovery: Identifies known or novel cell markers in complex tissues.
• Cell type composition analysis: Determines cell type proportions in dynamic tissue ecosystems.
• Cell-specific expression profiling: Estimates molecular expression profiles of distinct cell types from mixed tissue samples.

Methodology:

CAM3.0 implements three algorithms: radius-fixed clustering to identify reliable markers, linear programming to detect an initial scatter simplex, and a smart floating search for the optimum latent variable model.