neurips2022_multiome

Processed and subsampled of the data provided in the Multimodal Single-Cell Integration NeurIPS 2022 challenge (https://www.kaggle.com/competitions/open-problems-multimodal/data). <br> Data was filtered on donor "31800" and non-hidden celltypes. Subsequently, 2000 data points were randomly subsampled. The 2000 most highly variable genes were selected for the RNA data and peaks which appeared in less than 5% of the cells were filtered out, resulting in 11607 peaks.

本数据集基于NeurIPS 2022多模态单细胞整合挑战赛（Multimodal Single-Cell Integration NeurIPS 2022 challenge）提供的原始数据进行处理与二次抽样，原始数据获取链接为https://www.kaggle.com/competitions/open-problems-multimodal/data。
预处理阶段首先筛选出供体（donor）编号为"31800"的样本，同时保留所有非隐藏细胞类型的数据。
随后从筛选后的数据集内随机抽取2000个数据点作为样本子集。
针对转录组（RNA）数据，选取了变异程度最高的前2000个高可变基因（highly variable genes）；同时过滤掉仅在不足5%的细胞中出现的测序峰（peaks），最终共得到11607个测序峰。