DIPG和 adult glioblastomas (GBMs) 很大区别，不应该用同样的治疗方式。

测序策略是：

We integrated deep sequencing analysis of 36 tumor-normal pairs (20 whole-genome sequencing (Illumina HiSeq 2000) and 16 whole-exome sequencing (Applied Biosystems SOLiD 5500xl)) with comprehensive methylation (28 DIPGs; Illumina Infinium450k methylation array), copy number (45 DIPGs; Affymetrix SNP6.0) and expression (35 DIPGs; Illumina HT-12 v4) data (Supplementary Table 1). 

数据下载地址是：

Whole-genome sequencing data are accessible through the European Genome-phenome Archive (accession EGAS00001000575). 这个需要申请才能下载

Methylation data are accessible through the Gene Expression Omnibus (accession GSE50022).

Gene expression data are accessible through the Gene Expression Omnibus (accession GSE50021).

SNP6.0 copy number data are accessible through the Gene Expression Omnibus (accession GSE50024).

结果说明甲基化水平是最好的分组方式：(前面的文章提到过)

(a) Heat map of methylation levels in three DIPG subgroups identified by unsupervised hierarchical clustering. (b–d) Subrouping was supported by principal-components analysis

(b), non-negative matrix factorization (cophenetic coefficient = 0.9934, k = 3)

(c) and consensus clustering represented by a cumulative distribution function (CDF) and change in Gini coefficient

(d). PC, principal component.

分组是：

To understand what drives DIPGs, we integrated whole-genome sequencing with methylation, expression and copy number profiling, discovering that DIPGs comprise three molecularly distinct subgroups (H3-K27M, silent and MYCN) and uncovering a new recurrent activating mutation affecting the activin receptor gene ACVR1 in 20% of DIPGs

样本来源是：

Samples for 20 of these patients represented pretreatment samples (2 non-treated patients from autopsy), and 54 samples represented post-treatment autopsy samples. The median age of diagnosis was 6.37 years, with a median survival time of 10.4 months.

不同分组的特性是：

The MYCN subgroup had no recurrent mutations but was instead characterized by hypermethylation,
high-grade histology and chromothripsis on chromosome 2p leading to recurrent high-level amplification of MYCN and ID2
Tumors in this group overexpressed MYCN by 4-fold and 8-fold and
ID2 by 2.5-fold and 5-fold relative to the H3-K27M and silent groups, respectively.
The top most overexpressed genes in the MYCN group included FAP, HRSP12 and DYX2.
Therapies aimed at targeting altered histone modifications would not be effective in this subgroup.
Rather, children with this subtype of DIPG will potentially benefit from therapies targeting MYCN or possibly ID2.
Tumors in the silent subgroup had silent genomes on the basis of both whole-genome sequencing, structural and SNP6.0 copy number analysis and had a lower mutation rate than tumors in the other two subgroups
H3-K27M subgroup DIPGs were highly mutated in either histone H3.3 (H3F3A) or H3.1 (HIST1H3B and HIST1H3C).
This group had highly unstable genomes
(segmentation analysis of SNP6.0 data; 497 CNAs per genome versus 300 CNAs per genome in the silent group; P = 0.04).
TP53 mutations were enriched in this group
Similarly, PVT1, MYC and PDGFRA gains or amplifications and structural variants.
Given the complexity and heterogeneity of genetic changes for the H3-K27M subgroup, if therapies targeting the histone mutations become available, this subgroup would likely require multimodal therapies.
After H3F3A and TP53, the next most frequently mutated gene in DIPG was ACVR1 (encoding activin A receptor, type I), a new cancer gene.
Mutations of ACVR1 in four DIPGs (c.617G>A) resulted in a p.Arg206His substitution.