Female scientist holding sample that turns into genetic code
Dec. 2023 | Genomics

How to avoid wasting RNA-seq reads in metatranscriptomics

“Why am I not getting ANY usable RNA-seq reads in my metatranscriptomics data?!”
Stressed-out scientist

If you’ve ever Googled how to get high-quality reads in your metatranscriptomics data, then this blog post is for you.

Metatranscriptomics is complicated, but its rewards are enormous. It provides real-time information on the microbiome, revealing multiple details about microbial gene expression in diverse environments – from the soil to the gut. By focusing on what genes are expressed, metatranscriptomic analysis allows the characterization of the active functional profile of the entire microbial community.

However, you can run into issues with sensitivity when performing RNA-seq of complex microbial community samples. The consequence? The absence of on-target reads and the failure to capture low-abundance mRNA transcripts. This means hours upon hours of RNA-seq optimization to get it right.

Let’s dive into one of the most common but overlooked reasons for poor RNA-seq data in metatranscriptomic applications.

The culprit: Highly abundant ribosomal RNA

Ribosomal RNA (rRNA) is the primary villain in your RNA-seq story. It’s unwelcome but always present. It is over-represented unless reliably removed. Due to its sheer abundance but low scientific value, rRNA can compromise your ability to detect low-abundance RNA transcripts of interest. You end up with wasted reads, wasted time and spiraling costs. Imagine the intriguing gene expression details that might be obscured and never revealed…

And that’s why reliable rRNA removal is a highly critical but often overlooked step in RNA-seq optimization.

Revealing transcriptomic details you might otherwise miss

Not only does rRNA removal increase the sensitivity of your RNA-seq, but it also decreases the overall costs of generating transcriptome-wide data. It allows you to detect low-expressed transcripts with startling precision.

Efficient rRNA removal is especially significant in bacterial RNA-seq as bacterial RNAs typically lack poly(A) tails that are used for mRNA enrichment. Prokaryotic RNA profiling presents a unique challenge for rRNA removal due to the incredible diversity in rRNA sequences. For complex bacterial samples, such as in metatranscriptomics studies, hundreds of thousands of 5S, 16S and 23S sequences need to be removed. While different methods to remove bacterial rRNA are commonly used during RNA-seq library prep, these methods are not designed to be pan-bacteria. They exclusively focus on removing 16S and 23S sequences for a handful of common bacterial species and ignore the 5S rRNA.

A game changer in pan-bacterial rRNA removal

So, how do you eliminate all that unwanted pan-bacterial rRNA? How much effort will it require? Well, the good news is that there is a technology out there that can be a total game changer in pan-bacterial rRNA removal: QIAseq FastSelect.

QIAseq FastSelect –5S/16S/23S Kits have been developed specifically to remove not only 16S and 23S rRNA but also 5S rRNA from fragmented or full-length RNA for metatranscriptomics studies. By rapidly and efficiently removing a broad spectrum of bacterial rRNA during RNA-seq library prep, QIAseq FastSelect –5S/16S/23S Kits represent an optimal solution for both single-species isolates and complex community samples for metatranscriptomics analysis.

Fast is its middle name – literally

The name says it all: QIAseq FastSelect technology is super fast. With QIAseq FastSelect –5S/16S/23S Kits, you can achieve up to 95% removal of bacterial rRNA using a simple 14-min protocol, followed by bead cleanup. No need for hybrid capture or enzymatic digestion.

Show me the data!

Regardless of the bacterial species, as shown in Figure 1, QIAseq FastSelect –5S/16S/23S effectively removes rRNA, significantly reducing the proportion of reads allocated to rRNA.

Figure 1. Robust rRNA removal from gut community sample using QIAseq FastSelect. Total RNA was isolated from Gut Microbiome Whole cell Mix (ATCC) using the RNeasy PowerMicrobiome Kit (QIAGEN). Stranded transcriptome libraries were then prepared from 100 ng aliquots of the RNA using the QIAseq Stranded Total RNA Lib Kit (QIAGEN). For QIAseq FastSelect rRNA removal, QIAseq FastSelect reagent was added to the RNA sample, followed by fragmentation (89°C for 8 min) and stepwise cooling from 75°C to 25°C over 14 minutes. Following a bead cleanup, the remaining library prep steps were completed (starting with first strand synthesis). Sequencing was performed on a NextSeq 550, and data analysis was performed using CLC Genomics Workbench. [A] QIAseq FastSelect –5S/16S/23S results in highly efficient removal of rRNA. The percentage of reads mapped to rRNA is shown for both untreated and QIAseq FastSelect-treated samples, with QIAseq FastSelect removing nearly 87% of all rRNA. [B] The percentage of total reads mapped to rRNA is shown for “all bacteria”, as well as individual species.
Case studies highlighting the diverse applications of QIAseq FastSelect

How about some publications to back up those performance claims? Well, here are some highlighted metatranscriptomics studies demonstrating the efficacy of the QIAseq FastSelect –5S/16S/23S Kits, enough to convince the most cynical of scientists.

A study by Carbonne et al. was aimed at investigating how microbial communities from cheeses contribute to the development of organoleptic properties (1). Metatranscriptomic analyses were carried out to obtain a holistic functional profile of these microbial communities. An evaluation of the efficiency of RNA extraction from various cheese types and an assessment of mRNA enrichment procedures for metatranscriptomic analyses was conducted. For total RNA extracts from four cheeses, approximately 99% of the sequencing reads corresponded to rRNA. Following rRNA removal using QIAseq FastSelect, reads attributed to rRNA were significantly decreased.

Another example of the impact of QIAseq FastSelect is a study conducted by Yap et al., where different sequencing-based methodologies to distinguish viable from non-viable cells in a bovine milk matrix were evaluated (2). For the metatranscriptomics part of the study, QIAseq FastSelect was used to achieve successful rRNA removal before library prep and sequencing using Illumina and Oxford Nanopore technologies.

QIAseq FastSelect was also successfully applied in a comprehensive multi-omics study aimed at designating optimal reference strains representing the infant gut bifidobacterial species (3). Effective rRNA removal was achieved using the QIAseq FastSelect –5S/16S/23S Kit prior to RNA-seq of whole transcriptome libraries. The rRNA removal step optimized the samples for subsequent RNA-seq, allowing the assessment of microbe-microbe and microbe-host interactions.

The expression says it all…
“Yay! Finally, some on-target reads”
Smiling scientist

Now that you’ve mastered the art of maximizing gene expression reads from RNA-seq, you’re ready to conquer the murky waters of metatranscriptomics. Diving into a sea of sequences, who knows what intriguing details you might uncover about the secret lives of microbes…

Further resources of interest