Wastewater AMR surveillance: methods and regulatory readiness for SE Asia utilities

Wastewater is becoming a recognised public-health signal for antimicrobial resistance. What samples, what methods, and how do you make a deliverable that holds up to a regulator? A guide for utilities, public-health agencies, and environmental-health researchers in Southeast Asia.

Why wastewater AMR is gaining traction

Antimicrobial resistance (AMR) is one of the few public-health threats that is genuinely global. The World Health Organization has framed AMR as a top-tier global health priority, and Singapore (alongside many other jurisdictions) has placed AMR among its national strategic public-health priorities. Wastewater surveillance is one of several tools being explored to support this work.

Wastewater can be a useful population-level AMR signal at relatively low marginal cost. A sewer catchment integrates the antibiotic-use and microbial signature of every household, hospital, and clinic that drains into it. Where individual surveys can take months and significant resourcing, a wastewater grab sample can capture a community-level signal within a single sequencing run.

The WHO Tricycle protocol formalised a methodology for E. coli ESBL surveillance in wastewater and has been adopted by various utilities and research groups internationally. Beyond Tricycle, broader metagenomic AMR surveillance (capturing the full resistome rather than a single indicator) is increasingly used by groups that want flexibility and a defensible data trail. Whether either approach fits a given utility depends on the specific surveillance question, regulatory expectations, and budget.

What samples, where, and how often

A surveillance programme typically samples three points along the treatment train. Influent (raw sewage entering the plant) reflects community antibiotic use and the pathogens being shed. Activated sludge characterises the microbial reactor itself, important for treatment-performance questions. Effluent (treated water leaving the plant) tells you what is being released to the receiving environment.

Sampling cadence depends on the question. For population-level AMR surveillance, weekly grab samples at the influent are sufficient to track week-to-week trends and seasonality. For treatment-process performance, weekly composites across all three points are typical. For event-driven surveillance (outbreak response, regulatory submission) daily sampling for 1–2 weeks captures the relevant window.

Sample handling matters. AMR-gene profiles can drift within hours at room temperature. Best practice: sample on ice, ship cold-chain to the laboratory within 24 hours, freeze on arrival. Skip these steps and the resulting profile reflects more about your shipping logistics than the wastewater.

The core question: which AMR genes circulate in this catchment?

A useful AMR surveillance deliverable answers four operational questions. Which resistance gene families are present in this catchment? At what relative abundance? On which mobile genetic elements (plasmids, integrons), and therefore how transferable? And how does this change over time?

Each of these maps to a specific bioinformatics output. Resistance gene families come from sequence-based annotation against curated public resistance-gene databases. Relative abundance comes from read-mapping or assembly-based quantification. Mobile genetic elements come from co-localisation analysis on assembled contigs. Time series come from running the same pipeline on samples collected on a fixed cadence.

See our AMR surveillance dashboard for a live example, built for an academic-public-health partner tracking AMR across the Philippines.

Methods: shotgun metagenomics vs targeted AMR panels

Shotgun metagenomic sequencing is the more flexible and more comprehensive approach. Total DNA is extracted from the sample, sheared, and sequenced on an Illumina or PacBio instrument. Reads are then assembled into contigs and annotated against AMR databases. This captures the full resistome, including novel and divergent genes that targeted panels would miss.

Targeted AMR panels (qPCR arrays, amplicon-based gene panels) detect only the specific genes the panel was designed for. They are appropriate for routine monitoring of a defined set of priority genes (e.g., the WHO critical-priority list) but cannot answer "what new genes are appearing".

A common starting point for a utility starting from scratch is a metagenomic baseline (single timepoint, all sample sites) followed by routine targeted panels for ongoing surveillance, with metagenomics revisited periodically or when an unusual signal appears.

Reference databases: why one is not enough

No single AMR reference database is canonical. Each curates a different scope of resistance determinants and resolves at a different taxonomic granularity. Working bioinformaticians query several in parallel and reconcile the results.

The main public databases differ in scope. One is broad, with strict curation of how each gene confers resistance. Another focuses on acquired resistance genes in pathogens, with strong support for E. coli and other Tricycle-relevant organisms. A third is built for metagenomes, with short, non-redundant sequences suited to read mapping. A fourth is aligned with regulatory reporting.

A defensible deliverable names one primary reference and adds a cross-reference table from the others for the priority genes. Reviewers, regulators and downstream analysts then have the full picture, without the report inheriting a single database's biases.

Regulatory readiness: documentation, methods, reproducibility

A surveillance deliverable that holds up to scrutiny (internal review, regulator queries, or external audit) typically rests on three documentation pillars. A methods document, a written specification of sampling, extraction, sequencing, bioinformatics, and reporting, detailed enough that an independent laboratory could reproduce the workflow. A reproducible pipeline: workflows with pinned tool versions, containerised, runnable on cloud or on-prem. And an audit trail, sample IDs, processing dates, software versions, and parameter logs, retained for a duration that matches the operator's regulatory requirements (which vary by jurisdiction and use case).

We deliver engagements with each of these in mind: a methods PDF, a versioned pipeline repository, and an audit log compatible with retention policies. The aim is to be ready when an operational surveillance programme is asked, internally or externally, to show its work.

Pilot path: 12-week proof of concept

A typical first engagement: a 12-week pilot programme at one wastewater treatment plant. Week 1: sampling protocol agreement, shipping logistics, sample-handling QA. Weeks 2–10: weekly grab samples at the influent, shipped cold-chain to the laboratory, sequenced, and processed through the pipeline. Weeks 11–12: methods writeup, time-series analysis, regulatory documentation, and stakeholder briefing.

A 12-week pilot at one plant is enough to (a) detect seasonality patterns in the AMR signal, (b) demonstrate week-to-week reproducibility, (c) identify the priority genes worth ongoing monitoring, and (d) produce a methods document and surveillance-readiness assessment.

Most pilots either upgrade to an ongoing programme or are paused for budget reallocation. Either way, the data and methods are yours.

Where Nova Lifetech fits

We design, run, and document wastewater AMR surveillance pipelines for utilities and public-health agencies. We can work with the operator on sampling design, partner with the sequencing provider, and deliver a documented bioinformatics workflow plus a methods report the agency can submit upstream.

Where a utility already has internal capacity, we can provide consultation only, pipeline review, database curation, regulatory documentation. Where capacity is growing, we can run end-to-end and transition the workflow to the operator over time.

Start an enquiry if you are scoping a wastewater AMR programme or want a second opinion on a pilot already in progress.

A note on the figures and claims here

Sampling cadences, sequencing depths, and pilot timelines in this article are reasonable starting frameworks but vary by site, regulatory context, and surveillance question. The reference databases and protocols referred to here (such as WHO Tricycle) are publicly maintained resources that you can review directly. The "right" mix of methods and methodology should be settled in scoping conversations rather than assumed from a guide.

↗AMR surveillance dashboard: a live project
↗Shotgun metagenomics service detail
↗NGS at work in industry: wastewater section
Why wastewater AMR is gaining traction
What samples, where, and how often
The core question: which AMR genes circulate in this catchment?
Methods: shotgun metagenomics vs targeted AMR panels
Reference databases: why one is not enough
Regulatory readiness: documentation, methods, reproducibility
Pilot path: 12-week proof of concept
Where Nova Lifetech fits
A note on the figures and claims here

NGS at work in industry: four sectors, one shared blueprint

NGS used to be a research expense. Now it is an operational tool. Across aquaculture, wastewater, human microbiome research, and agriculture, microbial sequencing is replacing best-guess practice with measured insight, protecting yield, reducing risk, and surfacing value that did not exist a decade ago.

Designing a microbiome study: depth, primer choice, and where expert advice pays off

Two design decisions set the ceiling on what your microbiome study can detect: how deep you sequence each sample, and which 16S region you target. Get them wrong and no clever analysis afterwards can recover what was missed. Here is how to think about both, and where the value of an experienced bioinformatics partner shows up.

Tell us what you are working on.

A short note about your samples and questions is enough to start. We reply with how we would approach it.