Identification of core and rare species in metagenome samples based on shotgun metagenomic sequencing, Fourier transforms and spectral comparisons.
ISME Commun
; 1(1): 2, 2021 Mar 24.
Article
en En
| MEDLINE
| ID: mdl-37938695
In shotgun metagenomic sequencing applications, low signal-to-noise ratios may complicate species-level differentiation of genetically similar core species and impede high-confidence detection of rare species. However, core and rare species can take pivotal roles in their habitats and should hence be studied as one entity to gain insights into the total potential of microbial communities in terms of taxonomy and functionality. Here, we offer a solution towards increased species-level specificity, decreased false discovery and omission rates of core and rare species in complex metagenomic samples by introducing the rare species identifier (raspir) tool. The python software is based on discrete Fourier transforms and spectral comparisons of biological and reference frequency signals obtained from real and ideal distributions of short DNA reads mapping towards circular reference genomes. Simulation-based testing of raspir enabled the detection of rare species with genome coverages of less than 0.2%. Species-level differentiation of rare Escherichia coli and Shigella spp., as well as the clear delineation between human Streptococcus spp. was feasible with low false discovery (1.3%) and omission rates (13%). Publicly available human placenta sequencing data were reanalysed with raspir. Raspir was unable to identify placental microbial communities, reinforcing the sterile womb paradigm.
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Idioma:
En
Revista:
ISME Commun
Año:
2021
Tipo del documento:
Article
País de afiliación:
Alemania
Pais de publicación:
Reino Unido