MapID-based quantitative mapping of chemical modifications and expression of human transfer RNA

Detection and quantification of tRNA chemical modifications are critical for understanding their regulatory functions in biology and diseases. However, tRNA-seq-based methods for modification mapping encountered challenges both experimentally (poor processivity of heavily modified tRNAs during reverse transcription or RT) and bioinformatically (frequent reads misalignment to highly similar tRNA genes). Here, we report "MapID-tRNA-seq" where we deployed an evolved reverse transcriptase (RT-1306) into tRNA-seq and developed "MapIDs" that reduce redundancy of the human tRNA genome and explicitly annotate genetic variances. RT-1306 generated robust mutations against m¹A and m³C, and RT stops against multiple bulky roadblock modifications. MapID-assisted data processing enabled systematic exclusion of false-positive discoveries of modifications which arise from reads misalignment onto similar genes. We applied MapID-tRNA-seq into mapping m¹A, m³C and expression levels of tRNAs in three mammary cell lines, which revealed cell-type dependent modification sites and potential translational regulation of the reduced mitochondrial activities in breast cancer.