| Full name | Absolute Quantification RNA-Seq |
| Summary | AQRNA-Seq enables absolute quantification of all small RNA species in a sample by providing a direct, linear correlation between sequencing read count and RNA abundance |
| Key findings | Stress-induced mycobacterial persistence revealed large variations in tRNA copy numbers, tRNA fragmentation, and tRNA modification location and abundance within and among samples. |
| Sequencing strategy | RNA-Seq |
| Raw data | https://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA579244; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE139936; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE159434 |
| Software/data repository | https://github.com/dedonlab/; https://github.com/dedonlab/aqrnaseq; https://github.com/dedonlab/general_aqrnaseq |
| Experimental protocol | Adapter ligation begins at the 3’-end, with two randomized nucleotides at the 5’ end of linker 1 to maximize the T4 ligase efficiency. Linker 1 is DNA to facilitate removal of unligated linker with RecJ, a single-stranded-DNA-specific 5’→3’ exonuclease, leaving the hybrid RNA-DNA product intact. A 50:1 excess of linker 1 resulted in >90% ligation efficiency. Ligated RNA can then be treated with AlkB to reduce levels of RT-blocking methyl modifications. After demethylation RecJ digestion is performed to remove unligated linker. RT is accomplished with a DNA primer complementary to linker 1 and the resulting cDNA is 3’-ligated to a custom DNA adapter (linker 2) using T4 DNA ligase. Linker 2 possesses a hairpin, a random hexamer sequence (splint to enhance cDNA ligation), and a downstream primer binding site for subsequent amplification, with ligation optimized to >97% at 50:1 linker excess. Excess linker 2 is removed with RecJ and PCR amplification is performed with primers complementing linker 2 and incorporating a standard Illumina anchor and barcode for subsequent sequencing. |
| Analysis protocol | The AQRNA-seq pipeline begins with paired end sequence assembly that integrates read1 and read2 sequences, obtains high quality insert sequences by cross validating read pairs, and removes artificial linkers. The abundance of each unique insert sequence is counted in every sample and annotated with the corresponding RNA. The resulting non-redundant reference library is then used to align forward and reverse sequencing reads, first separating uniquely aligned reads from reads matching multiple sequences. For closely related reference sequences, multiply-mapped reads are resolved by collapsing ambiguous read assignments into separate groups, with subsequent determination of the proportion of multiply-mapped reads (i.e., do they significantly alter the final read count for each RNA) and the cause of multiple mapping (e.g., highly similar sequences such as tRNA isodecoders). These considerations rationalize a decision to discard, average, or sum the read counts from multiple, closely related reference sequences. Finally, the read count for full-length sequences and fragments is tabulated from the curated set of mapped reads. Data are normalized either to the total number of reads in each barcoded sample or to an internal RNA standard to account for sample-to-sample variation in input RNA as well as variable sample pooling prior to sequencing. |
| Organism/cell line | Mycobacterium tuberculosis variant bovis BCG |
| Conditions | stress-induced, non-replicative, antibiotic-resistant state of persistence in tuberculosis |
| Approximate experimental time | 3-4 days |
| Starting RNA amount | 50 ng small RNA |
| tRNA expression | ++ |
| Base modifications | + |
| Charging status | - |
| tRNA processing and fragmentation | ++ |
| Citation | https://doi.org/10.1038/s41587-021-00874-y |
