High-throughput gene expression profilng of SMARCA4-mutated extra-cranial rhabdoid tumours (ECRT-SMARCA4), SMARCB1-mutated ECRT, ATRT and SCCOHT tumours by RNA sequencing

Purpose: Molecular characterization of ECRT-SMARCA4 tumours and their place within the constellation of ECRT-SMARCB1, ATRT and SCCOHT Methods: Total RNA was obtained from 72 fresh-frozen tumour samples using the Qiagen RNAeasy kit (Qiagen, Venlo, Netherlands), according to the manufacturer's procedure. Barcode Illumina compatible libraries were generated from 750 ng of total RNA for each sample using the TruSeq Stranded mRNA Library Preparation Kit (Illumina). Libraries were sequenced using the Illumina HiSeq 2500 platform. RNA-seq data pre-processing was performed using an in-house pipeline developed at the Curie Institute Bioinformatics Core Facility. Read mapping and counting were performed using STAR, version 2.5.3) and the hg19 version of the human reference genome. Results: Dimensionality reduction and hierarchical clustering algorithms applied to the transcriptomics dataset show that ECRT-SMARCA4 display molecular features intermediate between SCCOHT and ECRTS-MARCB1. Overall design: Seventy two tumour samples were analysed by RNA sequencing including 38 ATRTs (29 re-analyzed samples from Leruste et al., 2019), 19 ECRT_SMARCB1 (all re-analyzed from Leruste et al., 2019), 4 ECRT_SMARCA4 and 11 SCCOHT (10 re-analyzed from Loarer et al., 2015 which 3 are deposited in SRA SRP052896: SRX857513, SRX857516, SRX857514). All previously published and re-analyzed data included in our study (as detailed in the published_re-analyzed_data.xlsx) have been directly provided by the authors as fastq files. For those already published in Leruste et al. (2019), the data are now under submission in the dbGAP database (phs001915.v1.p1) as mentioned in the Leruste et al. (2019) publication. Therefore, the accession numbers for each sample are not yet available. For the data from Loarer et al. (2015), only 3 are publicly available in SRA database under the accession number SRP052896.

研究目的：对ECRT-SMARCA4肿瘤开展分子特征鉴定，并明确其在ECRT-SMARCB1、ATRT与SCCOHT肿瘤谱系中的定位。
研究方法：参照厂商标准操作流程，使用凯杰（Qiagen）RNAeasy试剂盒（荷兰芬洛市凯杰公司），从72份新鲜冰冻肿瘤样本中提取总RNA。采用TruSeq链特异性mRNA文库制备试剂盒（Illumina公司），以每份样本750 ng总RNA为起始材料，构建适配Illumina条形码的测序文库。随后使用Illumina HiSeq 2500平台完成文库测序。RNA测序（RNA-seq）数据预处理采用居里研究所生物信息学核心团队开发的内部分析流程。读段比对与计数分析采用STAR v2.5.3软件及人类参考基因组hg19版本完成。
研究结果：对转录组数据集应用降维和层次聚类算法后可见，ECRT-SMARCA4肿瘤的分子特征介于SCCOHT与ECRT-SMARCB1肿瘤之间。
实验整体设计：本研究通过RNA测序分析了72份肿瘤样本，其中包含38份ATRT样本（29份为Leruste等人2019年研究的重分析样本）、19份ECRT_SMARCB1样本（全部来自Leruste等人2019年研究的重分析数据）、4份ECRT_SMARCA4样本以及11份SCCOHT样本（其中10份来自Loarer等人2015年研究的重分析数据，其中3份已存入SRA数据库SRP052896，具体登录号为SRX857513、SRX857516、SRX857514）。本研究纳入的所有已发表重分析数据（详见published_re-analyzed_data.xlsx文件）均由原作者以fastq文件形式直接提供。其中Leruste等人2019年发表的数据目前已提交至dbGAP数据库（登录号phs001915.v1.p1），相关信息已在Leruste等人2019年的发表论文中说明，因此各样本的具体登录号暂未公开。Loarer等人2015年的数据中，仅有3份可在SRA数据库中通过登录号SRP052896公开获取。