Advancing open science in big pharma: Integrating LIMMA/VOOM into the nf-core differential abundance

RNA-seq analysis has established itself as a powerful tool in clinical research, enabling the comprehensive profiling of gene expression. Clinical studies, however, present a complex landscape due to the multitude of clinical parameters that need to be measured, often leading to incomplete paired-patient data sets. Limma/voom has shown advantages in handling complex incomplete paired-patient data, a common occurrence in clinical studies. The nf-core differentialabundance pipeline is a versatile workflow for testing differentially expressedgenes or proteins from count or intensity-based high-throughput studies. Currently, the workflow supports limma for intensity-based and DESeq2 for count-based input. In a collaborative project between Ardigen and Boehringer Ingelheim, we have extended the nf-core differentialabundance pipeline for limma/voom analysis with mixed models which promises to enhance the accuracy of gene expression profiling, particularly in studies with missing data, and could potentially lead to more reliable biomedical research outcomes. To ensure adherence to the highest standards of open-source community and regulatory compliance, we have incorporated comprehensive testing and documentation into the pipeline. This initiative aligns with the nf-core community’s commitment to meeting the stringent requirements set forth by regulatory authorities.