Transcriptomics of mononuclear cells of Meniere Disease (MD) patients

Total RNA was sequenced to a minimum of 40 million 150 bp paired-end reads (12Gb) per sample. Library preparation was performed using the NEBNext® UltraTM Directional RNA Library Prep Kit (New England BioLabs, Massachusetts, USA). RNA-seq was performed on a Novaseq 6000 (Illumina, California, USA) at the Novogene Cambridge Science Park (UK) installations.

The data files from the Novaseq 6000 sequencing platform are transformed to sequence reads by CASAVA base recognition (Base Calling). RSEM software package was used for estimating gene and isoform expression levels.  The FASTQ files were pre-processed with BBTools to remove adapters as described in the sequencing library documentation, to trim low-quality regions (discarding reads of quality lower than 26) and selecting reads with a minimum length of 135 nucleotides. Reads were aligned to the GRCh38 reference human genome assembly using STAR (version 2.5).

Differential expression analysis was performed using the ExpHunter Suite Bioconductor package, which used DESeq2 and edgeR packages. Genes are labelled as prevalent or possible differentially expressed genes (DEG), based on package results: if a gene is detected as differentially expressed by the two packages it is considered a prevalent DEG. On the other hand, if a gene is detected as differentially expressed by only one, it is considered a possible DEG. A gene is considered differentially expressed if it presents an adjusted p < 0.05 and absolute logFC ≥ 1. ExpHunter Suite performs score integration to obtain combined logFC and adjusted p-value/FDR values for each gene across all packages. The comparisons for this study were: MD patients versus controls, MDL patients versus controls, MDH patients versus controls, and MDL versus MDH patients.