NextGenAMR — a product by Vanguard Biotech Systems
How NextGenAMR is designed to work
NextGenAMR is designed to take a bacterial genome within its scope through a controlled workflow, organise resistance-related signals, keep the traceability of how they were obtained and generate structured outputs for a professional to review together with other tests and the relevant context.
In development · Not for clinical use
In development. The E. coli engine is being built and will be evaluated internally under a pre-specified protocol. An independent external validation protocol has been drafted. No performance results have been published. Not for clinical use.
Detecting genomic evidence is not, by itself, a clinical interpretation. The absence of a known marker does not prove susceptibility. Any output must be reviewed together with the clinical context and the applicable tests.
Input
Input (by design) — Whole-genome sequencing reads (paired-end FASTQ) of a single cultured isolate. Not primary samples, not metagenomics.
This website never invites uploads: do not send samples, sequences, identifiers or patient data through it.
The engine, stage by stage
One status per stage: Built (internal R&D) · In development · Designed. No stage describes a result.
- 01Built (internal R&D)
Verified acquisition
Checksum-verified reads.
- 02Built (internal R&D)
Read quality control
Versioned quality gates with reason codes.
- 03Built (internal R&D)
De novo assembly
From quality-filtered reads, with reproducible coverage down-sampling and no reference-based pre-filtering.
- 04Built (internal R&D)
Assembly quality and species
Completeness, contamination and species; unsupported inputs rejected or flagged.
- 05Built (internal R&D)
Mechanism detection
Known determinants, versioned reference databases.
- 06Built (internal R&D)
Mechanism features
Three mechanism classes, encoded separately.
- 07In development
Probability of resistance
Per-antibiotic models, in development; evaluation criteria pre-specified.
- 08Designed
Uncertainty and abstention
Designed at model level; not yet validated.
- 09Designed
Auditable report
Designed; from a schema-validated result.
Output and abstention
Output (by design) — For each antibiotic in the predictive panel, a probability of resistance with explicit uncertainty — or an abstention. Never a categorical S/I/R result.
Abstention — A deliberate non-answer instead of a forced estimate. Today, quality-control and species checks reject or flag unsupported inputs; abstention at model level is designed and not yet validated.
Provenance
Every processing batch carries a write-once record of what produced it. Model packages and result records will join the same chain once they exist.
What a result is designed to contain
What a NextGenAMR result is designed to contain: a quality-control verdict with reason codes, species confirmation, the resistance mechanisms detected, a probability of resistance with its uncertainty for each antibiotic in the predictive panel — or an abstention with its reason — and the provenance of every step. No results are shown on this website.
Limits
- NextGenAMR is not presented as an autonomous diagnostic system; it is designed to organise genomic evidence and generate outputs for professional review.
- NextGenAMR does not replace phenotypic antimicrobial susceptibility testing (AST). Genomic outputs must be interpreted together with AST, microbiological context and applicable procedures.
- NextGenAMR does not select treatments and does not generate autonomous therapeutic recommendations. Clinical decisions remain with qualified professionals.
- NextGenAMR is in development and is not intended for clinical use. It has no authorisation or declaration of conformity for clinical diagnostic use: no CE marking under the EU In Vitro Diagnostic Regulation (IVDR, Regulation (EU) 2017/746) and no FDA marketing authorisation. No ISO 13485 quality-management certification is claimed.