P431 Tip to correct the variation position error in applying long-read high-throughput sequencing technology for fungal identification

Abstract

Poster session 3, September 23, 2022, 12:30 PM - 1:30 PM Presently, long-read high-throughput sequencing technology has been started applying in the medical mycology field instead of common Sanger sequencing. With its capability to sequence nucleic acid among the large region compared to Sanger sequencing, this technology can detect and analyze both inter- and intra-species sequence variation in internal transcribed spacer (ITS), a fungal house-keeping region, resulting in high accurate fungal identification. However, to set up the long-read high-throughput sequencing technology for fungal identification is quite challenging because of several factors related to the accuracy including the variation position error. So, in this study, we summarize the tip to correct the variation position error in applying long-read high-throughput sequencing technology for fungal identification that we have learnt in the preliminary laboratory setting using the 8 clinical isolates of yeast: Candida albicans (n = 2), C. tropicalis (n = 1), C. glabrata (n = 1), Trichosporon asahii (n = 2), Pichia kudriavzevii (n = 1), Cryptococcus neoformans var. grubii (n = 1). Based on the recruited strains, we found that self-assembling reference sequence generated from raw data of reading by using an auto-program named Canu causes the size-inflated sequence, a larger size calculated as 22.83 ± 7.56% than it should be, resulting in the shift of variation position. This error can be corrected by the alignment process of the reference sequence with the known sequence, both size and position, prior to doing the raw read alignment. The advantage of this process could correct not only for position shifting caused by the analysis process but also the random error generated from nanopore system. To validate this correction protocol, more samples are needed for further study.