KazRNA-Pipe-GPU-Accelerated, Reproducible Nextflow Workflow for Integrated Bulk and Single-Cell Transcriptomic Profiling

Medet Ashimgaliyev, Beimbet Daribayev, Аinur Zhumadillayeva, Miras Mussabek, Danil V. Lebedev, Bakhyt T. Matkarimov, Nurislam Kassymbek

2026

Full text Open access: Gold Metadata (CSL-JSON)

DOI 10.3390/biomedinformatics6040061

Data confidence 70 medium

How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Reproducible high-performance computing pipelines for transcriptomic analysis are essential for population-scale precision oncology, yet most published workflows address only a single sequencing modality or lack benchmarking on underrepresented populations. We present KazRNA-Pipe, an open-source Nextflow v26.04.6 workflow processing both bulk and single-cell RNA sequencing (RNA-seq) within a unified Singularity-containerized environment with graphics processing unit (GPU) acceleration through the RAPIDS ecosystem. The pipeline was validated on 22 esophageal squamous cell carcinoma (ESCC) bulk RNA-seq samples from a Kazakhstan cohort (PRJNA608223) and on the largest public ESCC single-cell atlas (GSE160269). Quantification concordance across STAR, HISAT2, and Salmon reached Spearman ρ ≥ 0.90. BayesPrism deconvolution resolved cell-type proportions across all 22 samples.