Annotation-aware multidomain validation of NucS and MutS-MutL across 23,435 prokaryotic reference genomes
Year: 2026
Type: article
Abstract: Background EndoMS/NucS mediates a non-canonical mismatch-repair pathway in several archaeal and actinobacterial lineages. A 2017 phylogenomic survey reported that NucS and canonical MutS-MutL were largely mutually exclusive, with coexistence concentrated in Halobacteria and Deinococcus-Thermus. The present study reassessed this distribution in a substantially expanded reference-genome collection while distinguishing annotation labels from multidomain protein-profile evidence. Results GFF3 annotations from 23,435 NCBI reference genomes (22,606 Bacteria and 829 Archaea) were screened, exact RefSeq protein accessions for annotation-positive candidates were retrieved, and candidate architectures were evaluated with Pfam profile hidden Markov models. A two-domain NucS call required NucS_N (PF21003) and NucS_C (PF01939) on the same candidate protein. A canonical MutS call required five MutS profiles (PF01624, PF05188, PF05192, PF05190 and PF00488) on the same protein without a same-protein PF01713 (SMR) warning. A MutL call required PF01119 and PF08676; PF13589 was supplementary. Annotation-only screening yielded 1,145 apparent coexistence genomes. Requiring two-domain NucS reduced this set to 473, and the combined NucS/MutS/MutL criteria yielded 470 genomes. Of these, 421 were Halobacteria, 48 were Deinococcota and one was a Lysobacter korlensis assembly with a severe contamination warning. The screen recovered two-domain NucS in 5,367 genomes. Explicit nucS gene annotations showed 99.63% concordance with the two-domain criterion (5,361/5,381), whereas only 0.73% of genomes classified solely from generic “NucS domain-containing” annotations passed both NucS profiles (5/682). Conclusions Annotation-only discovery markedly inflated apparent two-domain NucS occurrence and apparent NucS/MutS-MutL coexistence. Profile validation recovered a taxonomically concentrated coexistence pattern dominated by Halobacteria and Deinococcota, consistent with the qualitative pattern reported in 2017 despite the much larger reference collection. Because annotation-negative genomes were not exhaustively assessed by translated-genome searches, the profile-positive counts represent validated positives recovered by the annotation-first workflow rather than exhaustive prevalence estimates, and non-detection is treated as a screening outcome rather than evidence of biological gene loss.
Source: BMC Genomics
Author: Kiarash Sadeghian Esfahani
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $3,090
OpenAlex ID: W7204898840