Supplementary Figure 10: Low-input DART-Seq.

(a) Venn diagrams showing overlap among methylated RNAs identified by DART-Seq from sequencing libraries prepared using 1μg, 100ng, and 10ng total RNA as input (n = 3, 1, and 1 independent samples, respectively). (b) IGV browser tracks showing read coverage and C to U mutations in three representative mRNAs (BSG, EEF2, and HSPA5). C to U mutations occurring with at least 10% frequency are indicated by blue/red (C/U) coloring for the BSG transcript (positive strand) and by gold/green (C/U) coloring for the EEF2 and HSPA5 transcripts (negative strand). Tracks shown are for DART-Seq experiments using 1μg (n =3), 100ng (n =1), or 10ng (n =1) total RNA as input from cells expressing APOBEC1-YTH. The APOBEC1 track indicates DART-Seq reads from cells expressing APOBEC1 alone and prepared using 1μg total RNA as input (n = 2). MeRIP-Seq track is shown in blue to indicate m⁶A identified by antibody-based methods. (c) Low-input DART-Seq performs comparably to high-input DART-Seq in its ability to identify m⁶A sites. DART-Seq datasets from samples prepared using 10ng (n =1), 100ng (n =1), and 1μg (n =3) of total RNA as input were compared to m⁶A peak regions from three independent m⁶A immunoprecipitation datasets (m⁶A-Seq⁴, PA-m⁶A-Seq¹⁵, and MeRIP-Seq³). The proportion of m⁶A peaks from each dataset that overlaps with each of the three DART-Seq datasets is shown. (d) Motif analysis showing the most highly enriched motifs surrounding C to U sites identified by DART-Seq using 10ng (n = 73,205) or 100ng (n = 67,310) of total RNA as input. P values for individual motif enrichment were calculated using the cumulative binomial distribution. (e) Metagene analysis of C to U editing sites from HEK293T cells expressing APOBEC1-YTH and using 10ng of total RNA for sequencing (n = 1 sequencing sample).