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

Barcoded eGFP RNA reads with known poly(A) length were demultiplexed by locating the first 29 bases of eGFP sequence (see Table 2 ) within the first 250 bases of FASTA strings extracted from every FAST5 file. Next, the barcode was assigned by aligning the expected barcode sequences against the extracted read sequence preceding the eGFP alignment (see Table 2 ). The barcode with highest normalised alignment score (and above threshold of 0.6) was assigned to the read. To analyze barcoded eGFP DNA reads, the orientation of reads was investigated by aligning the first 29 bases of eGFP and its reverse-complement (Table 2) to the first 250 bases of FASTA strings extracted from each FAST5 file. A read was considered a poly(A)-containing read if the normalised alignment score of eGFP sequence was greater than both the normalised alignment score of the reverse-complement of eGFP and the threshold value of 0.5. Reads where the normalised alignment score of the reverse-complement of eGFP was higher than the forward eGFP sequence and passed the threshold value of 0.5 were considered to be poly(T)-containing reads. For Barcode demultiplexing, first the sequence preceding the identified eGFP start was queried for the presence of the experimentspecific PCR Front Primer in case of poly(A) reads, or its reverse-complement for poly(T) reads (sequences in Table 2 ).