NextGenAMR — a product by Vanguard Biotech Systems
For hospital laboratories
A single view for microbiology, validation, innovation, IT/security and institutional leadership: what NextGenAMR is designed to do, where its limits are, and how a validation collaboration with a laboratory that runs phenotypic AST would work.
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.
The context
WGS-based AMR analysis can combine several tools, databases, versions, rules, controls, reviews and reports. That composition is legitimate and powerful, but it can make consistency, versioning, provenance and systematic review harder to sustain over time.
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.
What NextGenAMR is
A genomic antimicrobial-resistance intelligence platform in development, designed to run a versioned workflow, apply a defined scope, record provenance, structure evidence, abstain when the applicable conditions are not met and produce results prepared for professional review. It is not autonomous diagnosis, not phenotypic AST, not a prescriber and not a treatment engine.
NextGenAMR is designed to be species-aware: each species has its own panel, features and models. The first species is Escherichia coli.
Input & scope
Input (by design) — Whole-genome sequencing reads (paired-end FASTQ) of a single cultured isolate. Not primary samples, not metagenomics.
- First species
- Escherichia coli — In development
- Candidate first panel
- 5 antibiotics, subject to evaluation
- Scope updated
- 2026-10-21
NextGenAMR's first species is Escherichia coli, in development, with a candidate first panel of ampicillin, ciprofloxacin, cefotaxime, ceftazidime and gentamicin, subject to evaluation. Carbapenems: genotypic evidence only, with no model probability. No species is yet released for use.
How it works →What it is designed to produce
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.
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.
Current 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.
- 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.
- NextGenAMR is in development; no performance results have been published.
- NextGenAMR is not presented as an autonomous diagnostic system; it is designed to organise genomic evidence and generate outputs for professional review.
- It does not replace phenotypic antibiogram (AST).
- It is designed to report probabilities of resistance, never a categorical S/I/R result.
- 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.
- Any output is for review by a qualified professional together with other evidence.
What a validation collaboration depends on
- The type and quality of the input, and the applicable organism and panel.
- The reference method (phenotypic AST) and a protocol agreed in advance.
- The transfer mechanism and the institutional approvals.
- The participation of the laboratory's scientific and technical leads.
Security & evidence at a glance
Security
How this website and the programme handle security, with each control's status taken from the internal register.
Security →Evidence & validation
Programme status: engine in development; internal evaluation criteria pre-specified; external validation protocol drafted.
Evidence & validation →Where each team looks
Microbiology & validation
Scope, output by design, abstention, evidence and programme status.
View →Innovation & operations
How a validation collaboration would work and what it does not mean.
View →IT, security & privacy
The security of this website and the programme, with each control's status.
View →Leadership & procurement
What NextGenAMR is today and the limits of a product in development.
View →