EhsanKA/sparsesampler_reproducibility: SparseSampler Reproducibility Code v1.0.0

Read full text
Authors: Ehsan Karimiara
Cited by: 0

Abstract

Reproducibility code for the SparseSampler (SPS) benchmarking study. What's included Sampling benchmark pipelines for SPS, Random, Hopper, Atomic Sketch, and scSampler Datasets: LCMV (flow cytometry, up to 34M cells) and MCC (scRNA-seq, up to 3.2M cells) with rare cell frequency variants (1%, 0.5%, 0.1%) EVR feature index sensitivity analysis (indices 1–30) Random Forest classification analysis comparing SPS vs Random sampling All figure generation scripts (coverage, runtime, UMAPs, EVR/RF/Rank analysis) Key parameters EVR Feature Index: 12 (default) Rare cell definition: centroid distance > 75th percentile AND frequency < threshold Reproducing results See README.md for step-by-step instructions covering data preprocessing, benchmark execution, and figure generation.

Fields of Study & Subjects

Single-cell and spatial transcriptomics Gene expression and cancer classification Cell Image Analysis Techniques

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
7 February 2026