A completeness-independent method for pre-selection of closely related genomes for species delineation in prokaryotes

Abstract

Background: Whole-genome approaches are widely preferred for species delineation in prokaryotes. However, these methods require pairwise alignments and calculations at the whole-genome level and thus are computationally intensive. To address this problem, a strategy consisting of sieving (pre-selecting closely related genomes) followed by alignment and calculation has been proposed.Results: Here, we initially test a published approach called "genome-wide tetranucleotide frequency correlation coefficient" (TETRA), which is specially tailored for sieving. Our results show that sieving by TETRA requires > 40% completeness for both genomes of a pair to yield > 95% sensitivity, indicating that TETRA is completenessdependent. Accordingly, we develop a novel algorithm called "fragment tetranucleotide frequency correlation coefficient" (FRAGTE), which uses fragments rather than whole genomes for sieving. Our results show that FRAGTE achieves~100% sensitivity and high specificity on simulated genomes, real genomes and metagenome-assembled genomes, demonstrating that FRAGTE is completeness-independent. Additionally, FRAGTE sieved a reduced number of total genomes for subsequent alignment and calculation to greatly improve computational efficiency for the process after sieving. Aside from this computational improvement, FRAGTE also reduces the computational cost for the sieving process. Consequently, FRAGTE extremely improves run efficiency for both the processes of sieving and after sieving (subsequent alignment and calculation) to together accelerate genome-wide species delineation.Conclusions: FRAGTE is a completeness-independent algorithm for sieving. Due to its high sensitivity, high specificity, highly reduced number of sieved genomes and highly improved runtime, FRAGTE will be helpful for whole-genome approaches to facilitate taxonomic studies in prokaryotes.

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

“…Horizontal gene transfer may yield a high ANI of ≥96% for even distantly related species, according to Additional file 4 in Zhou et al ( 2020 ). In this context, demarcating such species based merely on the ANI criterion may lead to incorrect conclusions.…”

“…However, almost all these methods except for FastANI suffer from a large amount of pairwise whole-genome comparisons. In an effect to address this obstacle, we proposed a strategy called “sieving,” the process of pre-selecting closely related (intraspecies and some closely related interspecies) pairs for subsequent genome-wide alignment and calculation (Zhou et al, 2020 ), and developed the method called “ frag ment te tranucleotide frequency correlation coefficient” (hereafter termed FRAGTE1) for it (Zhou et al, 2020 ). Our previous study shows that FRAGTE1 is completeness-independent, highly sensitive (~100%), and highly specific as well as with both highly reduced overall number of sieved pairs and highly improved runtime, compared with the “ tetra nucleotide frequency correlation coefficient” (TETRA), FastANI, and the alignment-based approaches such as 16S rRNA-based approach (Zhou et al, 2020 ).…”

“…However, the overall amount of pairs sieved by FRAGTE1 is still large, which is computationally intensive, especially for large datasets. Taking the dataset for real genomes consisting of 61,914 queries against 5,680 references used previously as an example, a total of 2,231,656 (0.63%) pairs were sieved (Zhou et al, 2020 ), leaving a room for developing a more robust sieving approach. Accordingly, here we developed an enhanced version of FRAGTE (hereafter termed FRAGTE2), which is tailored for sieving.…”

ANI, Mash and Dashing equally differentiate between Klebsiella species

“…(Richter and Rosselló-Móra, 2009). However, it is computationally intensive to calculate ANI, since genomic comparisons are based on pairwise alignments and calculations (Zhou et al, 2020). Apart from this, a recent revision has argued against the accuracy of the threshold proposed by the authors as a universal microbial species delineation, substantial sampling and species redundancy was found to alter the results and found no evidence of a universal genetic boundary among microbial species currently annotated in the NCBI taxonomy (Murray et al, 2021).…”

“…We found that both sketching programs, Mash and Dashing, performed well for draft genomes without even altering the cutoff for the complete genomes analyzed. Mash has been previously reported to also perform well in whole metagenome comparisons (Dong et al, 2020) and to be very useful in fungal taxonomy (Gostinčar, 2020), along with many other promising applications (Zhou et al, 2020).…”