depinwang/rnaseq-aligner-toy-benchmark-metrics-v2
收藏资源简介:
该数据集提供了五种RNA-seq比对工具(star、hisat2、subjunc、bwa-mem2、minimap2)在模拟GRCh38 RNA-seq读取上的性能评估指标。数据基于GENCODE v45转录本的真实内含子进行比对,评估指标包括每个(比对工具,样本)组合的剪接连接F1分数、精确度、召回率、真阳性、假阳性、假阴性等。数据集包含300行数据,每行代表一个比对工具在特定样本上的性能表现。数据来源包括rna-seq比对工具基准测试实验,使用Puhti (CSC)集群,参考基因组为GRCh38.primary_assembly加上GENCODE v45,模拟方法为uniform-sampled转录本且无错误模型,评分方法为精确元组匹配且最小读取数为3。
This dataset provides performance evaluation metrics for five RNA-seq aligners (star, hisat2, subjunc, bwa-mem2, minimap2) on simulated GRCh38 RNA-seq reads. The evaluation is based on ground-truth introns from GENCODE v45 transcripts, with metrics including per (aligner, sample) splice-junction F1 score, precision, recall, true positives, false positives, false negatives, etc. The dataset contains 300 rows, each representing the performance of an aligner on a specific sample. Data provenance includes the rnaseq-aligner-toy-benchmark experiment, Puhti (CSC) cluster, GRCh38.primary_assembly + GENCODE v45 reference, uniform-sampled transcripts simulation with no error model, and exact tuple match scoring with min_reads=3.




