Fig. 1: Overview of data preparation and Nanocompore steps.
From: RNA modifications detection by comparative Nanopore direct RNA sequencing
A Raw fast5 reads from 2 conditions are basecalled with Guppy, filtered with Samtools and the signal is then resquiggled with Nanopolish eventalign. The output of Nanopolish is then collapsed and indexed at the kmer level by NanopolishComp Eventalign_collapse. B Nanocompore aggregates median intensity and dwell time at transcript position level. The data is compared in a pairwise fashion position per position using univariate tests (KS, MW, t-tests) and/or a bivariate GMM classification method. The p-values are corrected for multiple tests and these data are saved in a database for further analyses. The signal graph is as an illustration not representative of all possible kmers.
