 |
 |
GEO help: Mouse over screen elements for information. |
|
| Status |
Public on Nov 05, 2018 |
| Title |
AK015 [tumor WGBS-seq] |
| Sample type |
SRA |
| |
|
| Source name |
adult glioblastoma tumour
|
| Organism |
Homo sapiens |
| Characteristics |
subtype: IDH
gender: male
age: 28
survival_status: alive
os (overall survival, month): 24
progression: 0
pfs (progression-free survival, month): 24
chr7_gain: 0
chr10_loss: 0
chr10q_loss: 0
chr19_gain: 0
chr20_gain: 0
idh1 mutation status: R132H
idh2 mutation status: wt
egfr_amplification: 0
pten_deletion: 0
mdm2_amplification: 0
mdm4_amplification: 0
pdgfra_amplification: 0
cdkn2a_b_deletion: 0
cdk4_amplification: 0
|
| Extracted molecule |
genomic DNA |
| Extraction protocol |
Nucleic acids were extracted from tumour tissue using Qiagen Allprep DNA/RNA/Protein Mini kit or a standard caesium chloride (CsCl) density gradient ultracentrifugation protocol. QC was performed using the Agilent Bioanalyser 2100 and the Nanodrop spectrophotometer.
Whole-genome bisulphite library preparation was carried out as described previously (see Hovestadt et al., Nature (2014) 510:537-541 https://doi.org/10.1038/nature13268). Briefly, 5 μg of genomic DNA was sheared using a Covaris device. After adaptor ligation, DNA fragments with insert lengths of 200–250 bp were isolated using an E-Gel electrophoresis system (Life Technologies) and the DNA was bisulphite converted overnight using the EZ DNA Methylation kit (Zymo Research). The fragments were PCR amplified using the FastStart High Fidelity PCR kit (Roche) for 6-8 cycles. Library aliquots were then purified and size selected with AMPure beads (New England BioLabs) and quality controlled with a Bioanalyzer (Agilent). Each library was sequenced using 2 lanes on an Illumina HiSeq 2000 in the DKFZ Genomics and Proteomics core facility.
|
| |
|
| Library strategy |
Bisulfite-Seq |
| Library source |
genomic |
| Library selection |
RANDOM |
| Instrument model |
Illumina HiSeq 2000 |
| |
|
| Description |
processed data file:
CR_analysis_full_table.txt
IDH_average_methylation_values.bigWig
IDH_methylation_features_DMV.bed
IDH_methylation_features_LMR.bed
IDH_methylation_features_PMD.bed
|
| Data processing |
For each sample, reads were mapped to the human genome (hg19) with bwa-mem (0.7.8) with a customized WGBS pipeline [ref methyCtools]. CpGs overlapping variable sites with a minor allele frequency higher than 0.25 were removed. Low coverage CpGs with 2 or fewer reads in more than 50% of the cohort were also removed from the analysis.
The mean methylation of the two cytosines in a CpG dinucleotide (one C on forward strand and the other on reverse strand) was calculated by weighting their CpG coverage, i.e. m = (m1*c1 + m2*c2) / (c1+c2), where m1 and m2 are the number of methylated CpGs of the two neighbouring cytosines. Similarly the mean coverage for the CpG dinucleotide is calculated by weighting the coverage itself: c = (c1*c1 + c2*c2)/(c1+c2) where c1 and c2 are the CpG coverage at corresponding cytosines.
(c1+c2) where c1 and c2 are the CpG coverage at corresponding cytosines.
Finally, the bsseq R package (1.10.0) was applied to smooth the methylation data and impute the missing methylation values with default parameters. CH methylation was also smoothed by bsseq.
Methylation features (LMRs, PMDs, DMVs) were called for each sample and consensus sets for each subtype generated (see the manuscript for details).
Correlating region analysis was performed to integrate methylation and ssRNAseq (see other Series) data for the GB tumours. CR detection was applied in a gene centric manner. For each protein coding gene, a sliding window of 6 CpG sites (maximum 10kb width) moves from 50kb upstream to 50kb downstream of the gene, with a step size of 3 CpG sites. The Spearman correlation and a correlation statistical test between mean CpG methylation (beta) in the window and the expression of the gene (TPM) was calculated. The windows are further filtered by correlation FDR <= 0.05 and maximum mean difference between subtypes >= 0.2. Negative correlated regions (negCRs) are defined as showing negative correlation between methylation and expression (i.e., high methylation corresponds to low expression) and positive correlated regions (posCRs) as the converse. Subtype specific CRs are defined as follows: the mean methylation in the subtype is higher or lower than all other subtypes, with a FDR < 0.05 from pairwise t-tests. We further analysed only the sets of subtype-specific CRs of at least 1000 regions.
Genome_build: hg19
Supplementary_files_format_and_content: deepTools2 was used to produce methylation values bigWigs, where each CpG methylation is shown on the beta scale [0,1]. Subtype average values are also included. Per-CpG methylation (beta) and read coverage files are also included in the following format: for each chromosome analysed (chr1:chr22), three files are included: CG_positions_chr.txt.gz (chromosomal coordinates of the CpG di-nt), CG_coverage_chr.txt.gz (per-sample coverage matrix) and CG_methylation_bsseq_smoothed_chr.txt.gz (per-sample methylation (beta) matrix). Each chr's files has the same number of rows, in the same order. Subtype methylation feature bedfiles showing the location of consensus subtype DMVs, LMRs and PMDs are included. Finally, the CR analysis table containing CR positions, target gene, CR region methylation for each sample (beta), correlation, and subtype-specificity is included with bedfiles containing positions of all neg/posCRs and subtype-specific CR sets.
|
| |
|
| Submission date |
Oct 24, 2018 |
| Last update date |
Nov 15, 2018 |
| Contact name |
Bernhard Radlwimmer |
| E-mail(s) |
b.radlwimmer@dkfz-heidelberg.de
|
| Organization name |
Deutsches Krebsforschungszentrum / German National Cancer Research Centre
|
| Department |
Department of Molecular Genetics
|
| Street address |
Im Neuenheimer Feld 280
|
| City |
Heidelberg |
| ZIP/Postal code |
69120 |
| Country |
Germany |
| |
|
| Platform ID |
GPL11154 |
| Series (2) |
| GSE121721 |
Glioblastoma epigenome profiling identifies SOX10 as a master regulator of molecular tumour subtype - tumour methylation (WGBS) data |
| GSE121723 |
Glioblastoma epigenome profiling identifies SOX10 as a master regulator of molecular tumour subtype |
|
| Relations |
| BioSample |
SAMN10285656 |
| Supplementary file |
Size |
Download |
File type/resource |
| GSM3444620_AK015_methylation_values.bigWig |
165.0 Mb |
(ftp)(http) |
BIGWIG |
| Raw data not provided for this record |
| Processed data provided as supplementary file |
| Processed data are available on Series record |
|
|
|
|
 |
