Frontiers in Microbiology (May 2022)

FRAGTE2: An Enhanced Algorithm to Pre-Select Closely Related Genomes for Bacterial Species Demarcation

  • Jiaqi Zeng,
  • Jiaqi Zeng,
  • Yuxiao Wang,
  • Ziyao Wu,
  • Yizhuang Zhou

DOI
https://doi.org/10.3389/fmicb.2022.847439
Journal volume & issue
Vol. 13

Abstract

Read online

We previously reported on FRAGTE (hereafter termed FRAGTE1), a promising algorithm for sieving (pre-selecting genome pairs for whole-genome species demarcation). However, the overall amount of pairs sieved by FRAGTE1 is still large, requiring seriously unaffordable computing cost, especially for large datasets. Here, we present FRAGTE2. Tests on simulated genomes, real genomes, and metagenome-assembled genomes revealed that (i) FRAGTE2 outstandingly reduces ~50–60.10% of the overall amount of pairs sieved by FRAGTE1, dramatically decreasing the computing cost required for whole-genome species demarcation afterward; (ii) FRAGTE2 shows superior sensitivity than FRAGTE1; (iii) FRAGTE2 shows higher specificity than FRAGTE1; and (iv) FRAGTE2 is faster than or comparable with FRAGTE1. Besides, FRAGTE2 is independent of genome completeness, the same as FRAGTE1. We therefore recommend FRAGTE2 tailored for sieving to facilitate species demarcation in prokaryotes.

Keywords