Additional file 2: of CellMapper: rapid and accurate inference of gene expression in difficult-to-isolate cell types
Year: 2016
Type: article
Abstract: Schematic of the CellMapper SVD filter and algorithm. CellMapper first performs an SVD of the microarray expression matrix to extract major components of variation (singular vectors). Then it re-weights the components of variation based on their estimated relevance to the query gene, with larger weights given to components that are tightly correlated with the query gene (e.g. â Component 3â is highly correlated with the query gene expression pattern and receives a large weight). Then the microarray data are reconstructed from the components using the estimated weights. The result of this SVD filter is to emphasize the components of variation that most distinguish the query gene and dampen components that are less relevant to the given query. After the SVD filtering process, genes are ranked based on the Pearsonâ s correlation of their transformed expression pattern to that of the query gene. (PDF 22 kb)
Source: Figshare
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7138335379