Datasets for 'Mandrake: visualising microbial population structure by embedding millions of genomes into a low-dimensional representation'

Datasets for the paper 'Mandrake: visualising microbial population structure by embedding millions of genomes into a low-dimensional representation' Files: 616k* - Files for the analysis of 661k bacterial genomes from the SRA (note typo 616-661k). Includes mandrake output and input files (.npz) gps_acc - Files for the analysis of 20k S. pneumoniae accessory genomes from the GPS project. Original accessory matrix is gps_gene_presence_absence.Rtab sc2million_v1* - Files for the analysis of ~1M SARS-CoV-2 genomes. sc2million_v3.npz are the input distances. sce<commit hash>.qdrep - Nvidia systems profile of code at that commit hash sce<commit hash>.ncu-rep - Nvidia kernel profile of code at that commit hash

本数据集为论文《Mandrake：通过将数百万个基因组嵌入低维表征以可视化微生物种群结构》的配套数据，包含以下文件组：
616k*：用于分析SRA数据库中66.1万个细菌基因组的分析文件（注：原文存在笔误，实际应为616k-661k），包含mandrake的输出与输入文件（格式为.npz）
gps_acc：用于分析GPS项目中2万个肺炎链球菌（S. pneumoniae）附属基因组的分析文件，其原始附属矩阵为gps_gene_presence_absence.Rtab
sc2million_v1*：用于分析约100万个SARS-CoV-2（严重急性呼吸综合征冠状病毒2）基因组的分析文件，其中sc2million_v3.npz为输入距离文件
sce<提交哈希值>.qdrep：该提交哈希值对应代码的Nvidia系统性能剖析文件
sce<提交哈希值>.ncu-rep：该提交哈希值对应代码的Nvidia内核性能剖析文件