,
Julie Devis [aut] ,
Anna Diacofotaki [ctb],
Charles De Smet [ths],
Laurent Gatto [aut, ths, cre] | Title: | Data companion to CTexploreR |
|---|---|
| Description: | Data from publicly available databases (GTEx, CCLE, TCGA and ENCODE) that go with CTexploreR in order to re-define a comprehensive and thoroughly curated list of CT genes and their main characteristics. |
| Authors: | Axelle Loriot [aut] ,
Julie Devis [aut] ,
Anna Diacofotaki [ctb],
Charles De Smet [ths],
Laurent Gatto [aut, ths, cre] |
| Maintainer: | Laurent Gatto <[email protected]> |
| License: | Artistic-2.0 |
| Version: | 1.5.0 |
| Built: | 2024-06-30 03:13:20 UTC |
| Source: | https://github.com/bioc/CTdata |
Correlation coefficients between Cancer-Testis genes and all genes found on the CCLE database.
A matrix object with 298 rows and 24327 columns.
Rows correspond to CT genes
Columns correspond to all genes from CCLE database
Correlation coefficients (Pearson) between CT genes and all other
genes are given in the matrix. These correlation coefficients were
calculated using log transformed expression values from CCLE_data
(all cell lines).
See scripts/make_CCLE_correlation_matrix.R for details.
Gene expression data in cancer cell lines from CCLE
A SummarizedExperiment object with 24327 rows and 1229 columns
Rows correspond to genes (ensembl_gene_id)
Columns correspond to CCLE cell lines
Expression data from the assay are TPM values
Cell lines metadata are stored in colData
The rowData contains
A column percent_of_positive_CCLE_cell_lines that gives the
percentage of CCLE cell lines (all cell lines combined)
expressing the gene at a highly level (TPM >= 10).
A column percent_of_negative_CCLE_cell_lines that gives the
percent of CCLE cell lines (all cell lines combined) in which
genes are completely repressed (TPM < 0.1)
A column max_TPM_in_CCLE that gives the maximal expression (in
TPM) found in all cell lines.
A column CCLE_category gives the category ("activated",
"not_activated", "leaky") assigned to each gene. "activated"
category corresponds to genes highly expressed (TPM >= 10) in at
least one cell line and repressed (TPM <= 0.1) in at least 20% of
cell lines. "not_activated" category corresponds to genes
repressed (TPM <= 0.1) in at least 20% of cell lines and never
expressed (TPM >= 10) in any cell line. "leaky" category
corresponds to genes repressed (TPM <= 0.1) in less than 20% of
cell lines.
TPM values downloaded using depmap bioconductor package (see
scripts/make_CCLE_data.R for details).
Cancer-Testis (CT) genes description
A tibble object with 298 rows and 36 columns.
Rows correspond to CT genes
Columns give CT genes characteristics
When the promoter is mentionned, it has been determined as 1000 nt upstream TSS and 200 nt downstream TSS.
CT_genes characteristics column:
Column family gives the gene family name.
Columns chr, strand and transcription_start_site give the
genegenomic location.
Column X_linked indicates if the gene is on the chromosome X
(TRUE) or not (FALSE).
Column TPM_testis gives the gene expression level in testis
(using GTEx database).
Column max_TPM_somatic gives the maximum expression level found
in a somatic tissue (using GTEx database).
Column GTEX_category gives the category ("testis_specific",
"testis_preferential" or "lowly_expressed") assigned to each gene
using GTEx database (see ?GTEX_data for details).
Column lowly_expressed_in_GTEX indicates if the gene is lowly
expressed in GTEX database and thus needed to be analysed with
multimapping allowed.
Column multimapping_analysis informs if the gene (flagged as
"lowly_expressed" in GTEX_data) was found to be testis-specific
when multi-mapped reads were counted for gene expression in
normal tissues ("not_analysed" or "testis_specific") (see
?normal_tissues_multimapping_data for details).
Column testis_specificity gives the testis-specificity of genes
assigned to each gene using GTEX_category and
multimapping_analysis ("testis_specific" or
"testis_preferential").
Columntestis_cell_type specifies the testis cell-type showing the highest
mean expression of each gene (based on testis scRNAseq data).
Column Higher_in_somatic_cell_type specifies if a somatic cell type
was found to express the gene at a higher level than any germ cell type
(based on scRNAseq data of different tissues).
Column percent_of_positive_CCLE_cell_lines gives the percentage
of CCLE cancer cell lines in which genes are expressed (genes
were considered as expressed if TPM >= 10).
Column percent_of_negative_CCLE_cell_lines gives the percentage
of CCLE cancer cell lines in which genes are repressed (TPM <=
0.1).
Column max_TPM_in_CCLE gives the highest expression level of
genes in CCLE cell lines.
Column CCLE_category gives the category assigned to each gene
using CCLE data. "Activated" category corresponds to genes
expressed in at least one cell line (TPM >= 10) and repressed in
at least 20% of cell lines.
Column percent_pos_tum gives the percentage of TCGA cancer
samples in which genes are expressed (genes were considered as
expressed if TPM >= 10).
Column percent_neg_tum gives the percentage of TCGA cancer samples in
which genes are repressed (TPM <= 0.1).
Column max_TPM_in_TCGA gives the highest expression level of
genes in TCGA cancer sample.
Column TCGA_category gives the category assigned to each gene
using TCGA data. "activated" category corresponds to genes
expressed in at least one tumor (TPM >= 10) and repressed in at
least 20% of samples. "multimapping_issue" corresponds to genes
that need multi-mapping to be allowed in order to be analysed
properly.
Column DAC_induced summarises the results (TRUE or FALSE)
of a differential expression evaluating gene induction upon DAC
treatment in a series of cell lines.
Column somatic_met_level that gives the mean methylation level
of each promoter in somatic tissues.
Column sperm_met_level that gives the methylation level of each
promoter in sperm.
Column somatic_methylation indicates if the promoter's mean
methylation level in somatic tissues is higher than 50%.
Column germline_methylationindicates if the promoter is
methylated in germline, based on the ratio with somatic tissues
(FALSE if somatic_met_level is at least twice higher than
germline_met_level).
Column regulated_by_methylation indicates if the gene is
regulated by methylation (TRUE) based on DAC induction (has to
be TRUE) and on promoter methylation levels in normal tissues
(when available, has to be methylated in somatic and unmethylated
in germline).
Column named CpG_density, gives the density of CpG within each
promoter (number of CpG / promoter length * 100).
Column CpG_promoter classifies the promoters according to their
CpG densities: "low" (CpG_density < 2), "intermediate"
(CpG_density >= 2 & CpG_density < 4), and "high" (CpG_density >=
4).
Columns external_transcript_name, ensembl_transcript_id, and
transcript_biotype give the references and informations about
the most biologically relevant transcript associated to each
gene.
Columns oncogene and tumor_suppressor informs if oncogenic
and tumor-suppressor functions have been associated to genes
(source: Cancermine).
See scripts/make_CT_genes.R for details on how this list of
curated CT genes was created.
Mean methylation values of all CpGs located within Cancer-Testis (CT) promoters in a set of normal tissues
A SummarizedExperiment object with 298 rows and 14 columns
Rows correspond to CT genes
Mean methylation levels in normal tissues are stored in columns
CpG densities and results of methylation analysis are stored in rowData
The rowData contains:
A column named CpG_density, gives the density of CpG within
each promoter (number of CpG / promoter length * 100).
A column CpG_promoter that classifies the promoters according
to their CpG densities: "low" (CpG_density < 2), "intermediate"
(CpG_density >= 2 & CpG_density < 4), and "high" (CpG_density >=
4).
A column somatic_met_level that gives the mean methylation
level of each promoter in somatic tissues.
A column sperm_met_level that gives the methylation level of
each promoter in sperm.
A column somatic_methylation indicates if the promoter's mean
methylation level in somatic tissues is higher than 50%.
A column germline_methylationindicates if the promoter is
methylated in germline, based on the ratio with somatic tissues
(FALSE if somatic_met_level is at least twice higher than
germline_met_level).
WGBS methylation data was downloaded from Encode and from GEO
databases. Mean methylation levels are evaluated using methylation
values of CpGs located in promoter region (defined as 1000 nt
upstream TSS and 200 nt downstream TSS) (see
scripts/make_CT_mean_methylation_in_tissues.R for details).
Methylation values of CpGs located within Cancer-Testis (CT) promoters in a set of normal tissues.
A RangedSummarizedExperiment object with 51725 rows and 14 columns
Rows correspond to CpGs (located within CT genes promoters)
Columns correspond to normal tissues
Methylation values from WGBS data
rowRanges correspond to CpG positions
WGBS methylation data was downloaded from Encode and from GEO
databases (see scripts/make_CT_methylation_in_tissues.R for
details).
This is the companion Package for CTexploreR containing omics
data to select and characterise CT genes.
Data come from public databases and include expression and methylation values of genes in normal and tumor samples as well as in tumor cell lines, and expression in cells treated with a demethylating agent is also available.
The CTdata() function returns a data.frame with all the
annotated datasets provided in the package. For details on these
individual datasets, refer to their respective manual pages.
See the vignette and the respective manuals pages for more details about the package and the data themselves.
CTdata()CTdata()
A data.frame describing the data available in
CTdata.
Laurent Gatto
CTdata()CTdata()
Gene expression values in a set of cell lines treated or not with 5-Aza-2'-Deoxycytidine (DAC), a demethylating agent.
A SummarizedExperiment object with 24359 rows and 32 columns
Rows correspond to genes (ensembl_gene_id).
Columns correspond to samples.
Expression data correspond to counts that have been normalised (by DESeq2 method) and log-transformed (log1p).
The colData contains the SRA references of the fastq files that were downloaded, and informations about the cell lines and the DAC treatment.
The rowData contains the results of a differential expression
evaluating the DAC treatment effect. For each each cell line, the
log2FC between treated and control cells is given, as well as the
p-adjusted value. The column induced flags genes significantly
induced (log2FoldChange >= 2 and padj <= 0.05) in at least one
cell line.
Differential expression analysis was done using DESeq2_1.36.0,
using as design = ~ treatment (see
scripts/make_DAC_treated_cells.R for details).
RNAseq
fastq files were downloaded from Encode database. SRA reference of samples are stored in the colData.
Gene expression values in a set of cell lines treated or not with 5-Aza-2'-Deoxycytidine (DAC), a demethylating agent. Many CT genes belong to gene families from which members have identical or nearly identical sequences. Some CT can only be detected in RNAseq data in which multimapping reads are not discarded.
A SummarizedExperiment object with 24359 rows and 32 columns
Rows correspond to genes (ensembl_gene_id).
Columns correspond to samples.
Expression data correspond to counts that have been normalised (by DESeq2 method) and log-transformed (log1p).
The colData contains the SRA references of the fastq files that were downloaded, and informations about the cell lines and the DAC treatment.
The rowData contains the results of a differential expression
evaluating the DAC treatment effect. For each each cell line, the
log2FC between treated and control cells is given, as well as the
p-adjusted value. The column induced flags genes significantly
induced (log2FoldChange >= 2 and padj <= 0.05) in at least one
cell line.
Differential expression analysis was done using DESeq2_1.36.0,
using as design = ~ treatment (see
scripts/make_DAC_treated_cells_multimapping.R for details).
RNAseq fastq files were downloaded from Encode database. SRA reference of samples are stored in the colData.
Gene expression data in normal tissues from GTEx database.
A SummarizedExperiment object with 24359 rows and 32 columns
Rows correspond to genes (ensembl_gene_id as rownames)
Columns correspond to tissues
Expression data from the assay are TPM values
The rowData contains
A column named GTEX_category, specifying the tissue specificity
category ("testis_specific", "testis-preferential",
"lowly_expressed" or "other") assigned to each gene using
expression values in testis and in somatic tissues, has been
added to the rowData. "testis_specific" genes are expressed
exclusively in testis (expression in testis >= 1 TPM, highest
expression in somatic tissues < 0.5 TPM, and expressed at least
10x more in testis than in any somatic
tissue). "testis-preferential" genes are genes expressed in
testis but also in a few somatic tissues (expression in testis >=
1 TPM, quantile 75% of expression in somatic tissues < 0.5 TPM,
and expressed at least 10x more in testis than in any somatic
tissue). "lowly_expressed" genes are genes undetectable in GTEX
database probably due to multi-mapping issues (expression in all
GTEX tissues < 1 TPM).
A column named max_TPM_somatic giving the maximum expression
level found in a somatic tissue.
Downloaded from
https://storage.googleapis.com/gtex_analysis_v8/rna_seq_data/GTEx_Analysis_2017-06-05_v8_RNASeQCv1.1.9_gene_median_tpm.gct.gz.
Some categories of tissues were pooled (mean expression values are
given in pooled tissues) (see scripts/make_GTEX_data.R for
details).
A short function that returns the default CTdata tags and, if provided, additional data-specific tags.
makeTags(x)makeTags(x)
x |
An optional |
A character containing the default tags and optional
data-specific tags. If x is missing or is of length 0, the
default tags are returned. Otherwise, a vector of length equal to
length(x) is returned.
CTdata:::makeTags() ## only default tags CTdata:::makeTags(character()) ## only default tags CTdata:::makeTags("myTag") ## one additional tag CTdata:::makeTags(c("myTag", "myOtherTag")) ## two additional tagCTdata:::makeTags() ## only default tags CTdata:::makeTags(character()) ## only default tags CTdata:::makeTags("myTag") ## one additional tag CTdata:::makeTags(c("myTag", "myOtherTag")) ## two additional tag
Gene expression values (TPM) in a set of normal tissues obtained by counting or not multi-mapped reads. Many CT genes belong to gene families from which members have identical or nearly identical sequences. Some CT can only be detected in RNAseq data in which multimapping reads are not discarded.
A SummarizedExperiment object with 24359 rows and 18 columns
Rows correspond to genes (ensembl_gene_id)
Columns correspond to normal tissues.
First assay, TPM_no_multimapping, gives TPM expression values
obtained when discarding multimapped reads.
Second assay, TPM_with_multimapping, gives TPM expression
values obtained by counting multimapped reads.
A column named multimapping_analysis has been added to the
rowData. It summarizes the testis specificity analysis of genes
flagged as "lowly_expressed" in GTEX_data. Genes are considered
"testis_specific" when, with multimapping allowed, they are
detectable in testis (TPM >= 1), their TPM value has increased
compared to without multimapping (ratio > 5), and their TPM value
is at least 10 times higher in testis than in any other somatic
tissue.
RNAseq fastq files were downloaded from Encode database (see
scripts/make_normal_tissues_multimapping.R for details).
Gene expression profiles in different human cell types based on scRNAseq data obtained from the Human Protein Atlas (https://www.proteinatlas.org)
A SingleCellExperiment object with 20082 rows and 66 columns
Rows correspond to genes (ensembl gene id as rownames)
Columns correspond to cell types
Expression values correspond to transcripts per million protein coding genes (pTPM)
Description of the colData:
Column Cell_type gives cell type.
Column group gives the cell type group (defined in the Human Protein Atlas).
Description of the rowData:
Column max_TPM_in_a_somatic_cell_type gives the maximum expression value
found in a somatic cell type
Column max_in_germcells_group gives the maximum expression value found
in a germ cell type
Column Higher_in_somatic_cell_type specifies if a somatic cell type
Gene expression values in cell types, based on multiple scRNAseq datasets
obtained from the Human Protein Atlas
(https://www.proteinatlas.org/about/download)
The data were converted in a SummarizedExperiment
(see scripts/14_make_scRNAseq_HPA.R for details).
Methylation values of probes located within Cancer-Testis (CT) promoters in samples from TCGA (tumor and peritumoral samples)
A RangedSummarizedExperiment object with 666 rows and 3423
columns
Rows correspond to Infinium 450k probes
Columns correspond to samples
Methylation data from the assay are Beta values
Clinical information are stored in colData
Probe information (hg38 coordinates) are stored in rowRanges
SKCM, LUAD, LUSC, COAD, ESCA, BRCA and HNSC methylation data were
downloaded with TCGAbiolinks and subsetted to select probes located
in CT genes promoter regions (see
scripts/make_TCGA_CT_methylation.R for details).
Gene expression data in TCGA samples (tumor and peritumoral samples).
A SummarizedExperiment object with 24350 rows and 4141 columns
Rows correspond to genes (ensembl_gene_id)
Columns correspond to samples
Expression data from the assay are TPM values
Clinical information are stored in colData
Genes information are stored in rowData
The colData contains clinical data from TCGA as well as global
hypomethylation levels obtained from paper DNA methylation loss
promotes immune evasion of tumours with high mutation and copy
number load from Jang et al., Nature Commun 2019 that were added
(see inst/scripts/make_TCGA_TPM.R for details).
The rowData contains genes information and, for each gene, the
percentage of tumors that are positive (TPM >= 10), and the
percentage of tumors that are negative (TPM < 0.1). In column
TCGA_category, genes are labelled as "activated" when the
percentage of positive tumors is > 0 and when at least 20% of
tumors are negative. Genes are labelled as "not_activated" when
the percentage of positive tumors is 0. Genes are labelled as
"leaky" when less than 20% of tumors are negative.
SKCM, LUAD, LUSC, COAD, ESCA, BRCA and HNSC expression data were
downloaded with TCGAbiolinks (see scripts/make_TCGA_TPM.R
for details).
Testis single cell RNAseq data from
The adult human testis transcriptional cell atlas (Guo et al. 2018)
A SingleCellExperiment object with 19777 rows and 6490 columns
Rows correspond to genes (gene names as rownames)
Columns correspond to testis cells
Description of the colData:
Column nGene gives the number of distinct genes detected per cell.
Column nUMI gives the total UMI number per cell.
Column clusters gives cluster number defined in the Guo's paper.
Column type gives the testis cell type associated to the cluster number.
Column Donor gives the Donor origin of the cells.
Description of the rowData:
Column percent_pos_testis_germcells gives the percent of testis germ cells
in wich the genes are detected (count > 0) (based on testis scRNAseq data).
Column percent_pos_testis_somatic gives the percent of testis somatic cells
in wich the genes are detected (count > 0) (based on testis scRNAseq data).
Columntestis_cell_type specifies the testis cell-type showing the highest
mean expression of each gene (based on testis scRNAseq data).
The rowData contains the testis_cell_type column, specifying the testis
cell-type showing the highest mean expression of each gene.
The count matrix GSE112013_Combined_UMI_table.txt.gz was downloaded from
GEO (accession: GSE11201). Metadata correspond to TableS1 from the paper's
supplemental data. The data were converted in a SingleCellExperiment
(see scripts/13_make_testis_sce.R for details).