Extensive studies have demonstrated that small RNAs—including miRNAs, piRNAs, and small interfering RNAs (siRNAs)—play essential regulatory roles in reproduction, development, epigenetic regulation, and tumorigenesis. The regulatory activity of small RNAs is closely associated with their intracellular abundance, and highly expressed small RNAs generally exert stronger repression of target genes. In addition, strand selection from double-stranded small RNA precursors is critical for small RNA function.
Therefore, accurate absolute quantification is essential for understanding small RNA biology. Quantification bias not only compromises the reliability of basic research but also hampers the discovery of disease biomarkers and the development of precision medicine.
In a study published in Nature Communications, the team led by Prof. WU Ligang from the Center for Excellence in Molecular Cell Science (Shanghai Institute of Biochemistry and Cell Biology) of the Chinese Academy of Sciences, developed a deep-sequencing based method termed 4NBoost, which enables highly accurate absolute quantification of small non-coding RNAs (sncRNAs), including microRNAs (miRNAs) and PIWI-interacting RNAs (piRNAs). Using this approach, the team constructed a quantitative atlas of small RNAs across diverse mammalian tissues, Arabidopsis thaliana tissues, and commonly used cell lines. In addition, they established the SmRNAQuant database ( http://wulg-lab.sibcb.ac.cn/SmRNAQuant/ ), which provides researchers with convenient and reliable access to high-precision small RNA expression profiles for public use and in-depth exploration.
The rapid development of next-generation sequencing technologies has greatly expanded small RNA research. However, mainstream library preparation methods rely on T4 RNA ligase-mediated adapter ligation. Due to the intrinsic structure-dependent ligation efficiency of this enzyme, substantial quantitative bias is introduced, leading to systematic overestimation or underestimation of specific small RNAs and limiting accurate functional interpretation. To reduce such bias, several improved protocols have been proposed, including the use of randomized adapters or elevated PEG concentrations, with representative methods such as 4N-seq, AQ-seq, IsoSeek, and NEXTflex.
However, these approaches are mainly optimized for small RNAs with a 3' terminal 2' -hydroxyl (2' -OH), and perform poorly for piRNAs and plant miRNAs that typically carry a 2' -O-methyl (2' -Ome) modification. Moreover, most existing methods do not incorporate unique molecular identifiers (UMIs) to correct PCR amplification bias, making absolute quantification difficult to achieve. In addition, current small RNA databases, including miRBase, MirGeneDB, miRmine, and DIANA-miTED, are constructed from sequencing data affected by systematic bias, and a quantitatively accurate reference database for small RNAs has remained unavailable.
To address these challenges, the research team developed the 4NBoost sequencing method, which minimizes ligation bias through the use of randomized adapters and optimized library construction conditions. By incorporating exogenous RNA spike-ins with known concentrations to generate standard curves, the method enables precise absolute quantification of small RNA molecules within each library. Using 4NBoost, the authors systematically profiled small RNA expression across 20 mouse tissues, 18 crab-eating macaque tissues, 24 commonly used cell lines, and 4 Arabidopsis thaliana tissues, generating the most comprehensive reference atlas for absolute quantification of mammalian small RNAs to date.
Compared with existing datasets, this atlas demonstrates improved accuracy in small RNA abundance estimation, strand selection analysis, and tissue-specific expression profiling.
Large volumes of conventional small RNA sequencing data have accumulated over the past two decades across different species, tissues, cell types, and disease conditions, representing an invaluable resource for small RNA research. However, how to effectively utilize these datasets remains a major challenge.
To overcome this limitation, the authors applied machine learning approaches to model and correct systematic biases in conventional sequencing data, enabling direct transformation of legacy datasets into more accurate quantitative expression profiles. This strategy significantly enhances the value of historical data while avoiding the cost and resource consumption associated with repeated sequencing. Based on the unbiased quantitative data generated by 4NBoost, the team further constructed the SmRNAQuant database, which integrates bias-correction models and provides a convenient platform for querying and analyzing high-precision small RNA expression profiles.
Taken together, this study establishes an integrated framework consisting of the 4NBoost technology, bias-correction models, and the SmRNAQuant database, providing important tools and resources for both fundamental small RNA research and translational applications. The work marks a transition in small RNA research from the era of relative expression analysis to absolute quantification.
Reference: https://www.nature.com/articles/s41467-026-68812-7
Appendix：
