tailfindr: Alignment-free poly(A) length measurement for Oxford Nanopore RNA and DNA sequencing

Maximilian Krause; Adnan M. Niazi; Kornel Labun; Yamila N. Torres Cleuren; Florian S. Müller; Eivind Valen

In this work we present tailfindr, a versatile R tool that allows estimation of poly(A) tail lengths from basecalled ONT long-read sequencing data from both native RNA and DNA sequencing approaches. We show that tailfindr is able to detect the poly(A) tail boundaries of in vitro transcribed eGFP RNA molecules and estimate their lengths based on read-specific raw data normalisation. For molecules with known poly(A) tails from 30 nt up to 150 nt the estimates match well with the expected lengths (Fig. 1D) , however the shortest poly(A) tail (10 nt) was estimated to have longer tails than expected. We believe that this bias can be explained by sample contamination of this RNA molecule during preparations, or by inefficient oligo-dT sequencing adapter ligation to poly(A) tail stretches at or below 10 nt. Consistent with the latter explanation we observed that the barcoded 10 nt RNA molecule was underrepresented in the RNA sequencing libraries compared to input quantities (data not shown). Overall, tailfindr correctly estimates poly(A) tail lengths of in vitro transcribed RNAs over a wide range of lengths.