Deconvolution analysis

Tao Wei; Jihwan Park; Heng Xu; Han Luo; Gyeong Dae Kim

Improve Research Reproducibility A Bio-protocol resource

Home
Protocols

Preprint

Deconvolution analysis

Last updated date: Mar 28, 2022 Views: 627 Forks: 0

An abbreviated version of this protocol was published in Science Advances

Characterizing dedifferentiation of thyroid cancer by integrated analysis

Download PDF

Ask a question

How to cite

Favorite

#-----------------------------------------------------------------------------------------------------------------------------------------------
### download GSE file
#-----------------------------------------------------------------------------------------------------------------------------------------------

if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")

BiocManager::install("GEOquery")

library(GEOquery)

mygeomat <- getGEO("GSE33630", GSEMatrix=TRUE)

dataMatrix<-mygeomat$GSE33630_series_matrix.txt.gz@assayData$exprs
rownames(dataMatrix)<-mygeomat$GSE33630_series_matrix.txt.gz@featureData@data$`Gene Symbol`
dataMatrix[1:5,1:15]

#-----------------------------------------------------------------------------------------------------------------------------------------------
### deconvolution using CELLCODE
#-----------------------------------------------------------------------------------------------------------------------------------------------

library(CellCODE)

# Marker file is a marker matrix file that can be a reference for bulk data deconvolution, composed of about 15 markers for each cell type.
# markers are composed of well-known markers and differentially expressed genes in each cell type.

# represent the cell type markers to "1", but others "0".

# example

	Fol	PTC	PTC2
TFF3	1	0	0
TPO	1	0	0
MT1G	1	0	0
CRABP1	1	0	0
SNHG9	1	0	0
FN1	0	1	0
NPC2	0	1	0
PERP	0	1	0
KRT7	0	1	0
HOPX	0	1	0
MDK	0	0	1
S100A13	0	0	1
NPW	0	0	1
TESC	0	0	1
MS4A4A	0	0	1

marker = read.table( "thyroid_cellcode.txt", sep = "\t", header = T)
rownames(marker)=make.names(marker[,1], unique=T)
marker=marker[,-1]
dim(marker)
head(marker)
marker[1:4,1:5]

grp <- c("ATC", "ATC", "ATC", "ATC", "ATC", "ATC", "ATC", "ATC", "ATC", "ATC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "PTC", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "NOM", "ATC")

dataMatrix=as.matrix(dataMatrix)

marker=as.matrix(marker)

grp=factor(grp)

colnames(dataMatrix)=grp

SPVs=getAllSPVs(dataMatrix , grp=grp, marker , method="mixed", plot=T, mix.par=0.3)

write.table(SPVs, file="GSE33630.txt", sep="\t", quote=F)

# plot using the ggplot2 or other plotting program

How to cite：

Readers should cite both the Bio-protocol preprint and the original research article where this protocol was used:

Wei, T, Park, J, Xu, H, Luo, H and Kim, G(2022). Deconvolution analysis. Bio-protocol Preprint. bio-protocol.org/prep1602.
Luo, H., Xia, X., Kim, G. D., Liu, Y., Xue, Z., Zhang, L., Shu, Y., Yang, T., Chen, Y., Zhang, S., Chen, H., Zhang, W., Li, R., Tang, H., Dong, B., Fu, X., Cheng, W., Zhang, W., Yang, L., Peng, Y., Dai, L., Hu, H., Jiang, Y., Gong, C., Hu, Y., Zhu, J., Li, Z., Caulin, C., Wei, T., Park, J. and Xu, H.(2021). Characterizing dedifferentiation of thyroid cancer by integrated analysis . Science Advances 7(31). DOI: 10.1126/sciadv.abf3657