5.0 Challenges and limitations of RNA-Seq method
RNA sequencing (RNA‑Seq) has become an essential tool for analysing differential gene expression (DGE) utilized in characterizing specific tissues,134 a development that has given room for deeper insight into the complexity of the protein-coding transcriptome than previously understood. Some studies have preferred RNA-Seq over the earlier developed high-throughput RNA analysis by microarrays due to its numerous advantages.17 One of such promising opportunities is the detection of gene fusions and differential expression of transcripts known to cause disease. Lee et al52 reported that dysregulation of long non-coding RNAs had been associated with development of some diseases such as cancer, myocardial infarction and diabetes. Other advantages include increased dynamic range of expression, measurement of focal changes (such as single nucleotide variants (SNVs), insertions and deletions), detection of rare and novel transcript isoforms, splice variants and chimeric gene fusions (that include genes and transcripts hitherto not identified).20
Despite the merits of the RNA-Seq, some limitations especially associated with the library preparation protocols can cause biases and overvaluation of results.27,135 Because sequencing is sensitive to the quantity of transcripts, this can make abundant mRNAs to be overly represented in RNA-Seq libraries and can have influence on the majority of the reads. These mRNAs are evaluated with low stochastic variability between samples and can be found significant by differential expression analysis (DEA). Conversely, transcripts low in abundance receive few reads, which can subject them to noise and reduce their chances of being selected.27
Another factor that influences the transcript detection in RNA-Seq experiments is the length of the transcript itself, because a longer transcript has a higher possibility of being detected in the library, it will be considered significant after DEA which can subsequently affect the functional annotation of significant genes.136,137Excitingly, evidences have proven that microarrays can outperform RNA-Seq in detecting small non-coding RNAs like microRNAs. Another limitation of the RNA-Seq is the errors that arise from the quality of the RNA, as significant percentage of the total RNA come from ribosomal RNA (rRNA) while a very low percentage come from the mRNA, as a result, special considerations must be made in the methods in order to either enrich the mRNA (polyA selection) or reduce the rRNA levels.17 It is possible to avoid some of these issues with proper study design, however, bias and inconsistency associated with adapter ligation, cDNA synthesis, and amplification could be primarily dependent on library preparation and pre-processing procedures.138 Therefore, standardized procedures for addressing the limitations of the RNA-Seq and ascertaining the accuracy, reproducibility and precision in a variant clinically important settings are pertinent to enable the adoption of RNA-Seq tests.