SEESAW quantification data for temporal gene expression across osteoblastogenesis (B6xCAST), n=9

Osteoblast cells mature from a mesenchymal stem cell pool to become cells capable of forming bone matrix and mineralizing this matrix. The goal of this study was to characterize temporal changes in the transcriptome across osteoblast maturation, starting with committed mesenchymal stem cell/ early pre-osteoblast stage through to mature osteoblasts capable of matrix mineralization. Methods: Enriched populations of pre-osteoblast-like cells were obtained from neonatal calvaria from B6xCAST mice expressing CFP under the control of the Col3.6 promoter. These cells were placed into culture for 4 days, removed from culture and subjected to FACS sorting based on the presence/absence of CFP expression. Cells expressing CFP were returned to culture, subjected to an osteoblast differentiation cocktail and RNA was collected at 2, 4, 6, 8, 10, 12, 14, 16 and 18 days post differentiation. Methods II: mRNA profiles for each time point were generated by next generation RNA sequencing, using an Illumina HiSeq 2000. Three technical replicates per sample were sequenced. Overall design: Gene expression in calvarial osteoblasts from neonatal B6xCAST-Col3.6 CFP mice at 9 time points post differentiation. File description: the R data files (.rda) provide outputs of the scripts in the mikelove/osteoblast-quant GitHub repo (July 2022, commit 01d96490), having run the fishpond package function importAllelicCounts() followed by minimal filtering. The `_counts.rda` files contain SummarizedExperiment objects with estimated count, TPM abundance, and effective length, but do not contain inferential replicates (bootstrap counts), although the transcript-level allelic counts object contains bootstrap mean and variances for every isoform, sample, and allele. The other two `.rda` files are summarized to gene level. The `_quant_dirs.tgz` files contain all the Salmon quantification data including bootstraps for the 9 time points. They are grouped into sets of three for convenience. The `CAST_EiJ.diploid.fa.gz` file provides the transcript sequences that were used for Salmon quantification. The `B6xCAST_discordant_global_AI.csv` file contains the same information as presented in Table S1 of Wu et al (2022). These are TSS-level results for 134 genes showing significant and discordant patterns within gene. The other 6 CSV files provide global and dynamic AI testing results at three levels of resolution: gene level, isoform level (txp), and TSS level where TSS within 50bp are combined into a single TSS-group. The significance cutoff is a q-value of 0.05. The code used for generating these results is provided in the GitHub repo: FennecFish/osteoblast-test.成骨细胞（Osteoblast）由间充质干细胞池分化成熟，成为能够合成骨基质并介导其矿化的细胞。本研究旨在刻画成骨细胞成熟过程中转录组的动态变化，覆盖从定型间充质干细胞/早期前成骨细胞阶段，直至具备基质矿化能力的成熟成骨细胞的完整过程。 实验方法：从表达受Col3.6启动子调控的CFP的B6xCAST小鼠的新生颅盖骨中分离得到富集的前成骨细胞样细胞群。将上述细胞体外培养4天后，从培养体系中回收，基于CFP的表达与否进行荧光激活细胞分选（FACS）。将表达CFP的细胞重新接种至培养体系，加入成骨细胞分化诱导培养液，并分别在分化后第2、4、6、8、10、12、14、16和18天收集RNA样本。 方法二：采用Illumina HiSeq 2000平台开展下一代RNA测序，生成各时间点的mRNA表达谱；每个样本设置3次技术重复并进行测序。 实验整体设计：分析取自新生B6xCAST-Col3.6 CFP小鼠颅盖骨的成骨细胞在分化后9个时间点的基因表达情况。 文件说明：本数据集提供的R数据文件（.rda格式）为运行mikelove/osteoblast-quant的GitHub代码仓库（2022年7月，提交ID 01d96490）中的脚本所得结果，具体为先执行fishpond软件包的`importAllelicCounts()`函数，随后进行最小化过滤。 带`_counts.rda`后缀的文件包含SummarizedExperiment对象，其中存储了估算的计数、TPM (Transcripts Per Million)丰度及有效长度，但未包含推断重复（即自举计数）；不过转录本水平的等位基因计数对象则包含了每个转录本亚型、样本和等位基因的自举均值与方差。其余两个.rda文件已汇总至基因水平。 带`_quant_dirs.tgz`后缀的文件包含所有Salmon定量数据，涵盖9个时间点的自举数据；为便于使用，这些数据被分为3组。`CAST_EiJ.diploid.fa.gz`文件提供了用于Salmon定量的转录本序列。 `B6xCAST_discordant_global_AI.csv`文件包含与Wu等人2022年发表的表S1完全一致的信息，该数据为134个基因的转录起始位点（TSS）水平分析结果，这些基因在基因内部呈现出显著且不一致的等位基因特异性表达（AI）模式。 其余6个CSV文件提供了三种分辨率水平下的全局及动态AI检测结果：基因水平、转录本亚型水平（txp），以及将50bp范围内的TSS合并为单个TSS组的TSS水平。显著性阈值为q值0.05。生成上述分析结果的代码已在GitHub仓库FennecFish/osteoblast-test中公开。