HPC-CleanSeq: A Tool for Contamination Removal in Big RNA-Seq Datasets
Author(s)
Date Issued
2025
Type
article
Volume
15546
Start Page
282
End Page
293
Abstract
RNA is a vital cellular molecule responsible for gene expression,
adapting the organism to various environments and developmental stages. RNA-
seq, a massively sequencing technique, enables the identification and quantifica-
tion of active genes in specific tissues or conditions. Nowadays transcriptomic
research projects often reach hundreds of gigabytes, positioning this analysis
within the realm of Big Data. However, the RNA-seq protocol is highly sensitive
to contamination, which can affect data quality and analysis outcomes. Contam-
inants are generally classified as either exogenous—such as bacteria, viruses, or
fungi from external sources—or endogenous, such as ribosomal RNA (rRNA)
sequences that are not part of the target sample. Although several tools exist for
cleaning transcriptomic data, most struggle to efficiently handle large datasets,
requiring extensive computational resources and significant processing time.
This paper introduces HPC-CleanSeq, a bioinformatics pipeline designed to au-
tomate contaminant removal in RNA-seq data on High-Performance Computing
(HPC) systems. At the core of HPC-CleanSeq is Centrifuge, a well-established
tool for identifying and classifying DNA or RNA sequences from complex sam-
ples. HPC-CleanSeq is especially suited for large-scale metagenomic and RNA-
seq studies, enabling researchers to quickly detect which organisms (e.g., bacte-
ria, viruses, fungi) are present in a biological sample. The pipeline offers an in-
tuitive interface, allowing users to configure settings, manage HPC scripts, and
visualize results locally. With HPC-CleanSeq, researchers can upload FASTQ
files, initiate contaminant removal, and obtain clean data without requiring spe-
cialized computational skills, making advanced RNA-seq analysis accessible to
a broader scientific community.