What question did this study set out to answer?

To develop and evaluate Presto, a high-performance R package that scales Wilcoxon rank-sum and auROC analyses to millions of single-cell observations.

May 29, 2019Open Access

Presto scales Wilcoxon and auROC analyses to millions of observations

View Full Paper

Ask AI

Bookmark

Authors

IKIlya KorsunskyBroad Institute ANAparna NathanHarvard University NMNghia MillardBroad Institute

Discussion

Member takes

Implication

Computational benchmarking demonstrates rapid Wilcoxon and auROC analysis in massive single-cell RNA-seq datasets, highlighting dramatic speedups over standard implementations.

Key Points

To develop and evaluate Presto, a high-performance R package that scales Wilcoxon rank-sum and auROC analyses to millions of single-cell observations.
Engineered an optimized algorithm for dense and sparse matrix architectures with direct compatibility for Seurat and Bioconductor SingleCellExperiment frameworks.
Benchmarked computation speeds against base R functions using a sparse dataset comprising 1,000,000 observations across 10 groups and 1,000 features.
Presto completed rank-sum and auROC analyses on 1 million observations in 17 seconds, compared to 6.4 hours required by base R implementations.

Cite This Study

Korsunsky et al. (2019) studied this question.

synapsesocial.com/papers/6a6d19fc2163a0a01bc2bb2e https://doi.org/10.1101/653253

View Full Paper

Ask AI

Bookmark

Implication

Computational benchmarking demonstrates rapid Wilcoxon and auROC analysis in massive single-cell RNA-seq datasets, highlighting dramatic speedups over standard implementations.

Presto scales Wilcoxon and auROC analyses to millions of observations

Discussion

Member takes

Implication

Key Points

Cite This Study

Presto scales Wilcoxon and auROC analyses to millions of observations

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider