Skip the pipeline plumbing.
Keep every parameter.
spcfy runs your FASTQ through a standardized, versioned cloud pipeline that checks every candidate taxon against multiple marker-matched reference databases, keeps every threshold visible, and assigns stable identifiers so runs stay comparable over time. The scientific decisions stay yours.
Get in touch ↓About 30 minutes. No commitment. EU hosted, GDPR compliant.
Metabarcoding outgrew what a single pipeline can hold together.
One new sample arrives after a project is effectively finished, and the re-run swallows work that was already done.
One incomplete or biased reference database, and a single top hit produces a biologically implausible result.
A pile of tools, scripts and environments is not a stable pipeline, and someone has to keep checking whether it still works.
Filter too hard and rare taxa disappear. Filter too soft and noise stays in. Change a parameter or a database between runs, and yesterday's dataset stops lining up with today's.
One standardized pipeline, from FASTQ to comparable results.
Upload FASTQ
Demultiplexed or multiplexed, paired-end or pre-merged: five supported entry points cover common lab outputs.
Multi-database consensus
SPARK scores every candidate taxon across the reference databases that match your marker - BOLD, NCBI GenBank, SILVA, UNITE, RDP or MIDORI2 - instead of blindly trusting a single top hit.
Versioned, comparable results
SPIN identifiers merge new runs into existing projects without re-clustering the historic dataset. Every result carries its pipeline version, database version and QC thresholds.
Infrastructure that gives you back time and control.
Control and transparency
Set read-length filters and singleton handling, choose OTU clustering or ESV denoising, and bring your own regional or custom reference databases.
Reproducibility
Every dataset carries a locked audit trail: pipeline version, database version, QC thresholds.
Comparability
SPIN identifiers keep OTUs comparable across runs, projects and labs, without re-clustering.
Time back for the science
Hand off routine processing, database cross-querying and report generation. Keep experimental design, QC decisions and interpretation.
Marker coverage
CO1, 16S, ITS2, 12S and PITS2, across animals, plants, fungi and bacteria.
Ready to see it on your data?
Book a live demoDon’t just take our word for it. Hear it from the experts.
A new preprint compares five leading COI metabarcoding pipelines, spcfy among them, against a ground-truth dataset built from more than 11,000 individually DNA-barcoded arthropod specimens, representing around 1,700 Barcode Index Numbers (BINs). Scientists at the Centre for Biodiversity Genomics conducted the comparison independently of the spcfy team.
Although this benchmark focused on COI metabarcoding, spcfy was built from the start as a marker-independent platform, staying closely compatible with the Barcode Index Number (BIN) framework, the gold standard for animal DNA barcoding and a key foundation for the COI research community.
With thanks to Sean Prosser, Ken Thompson, Paul Hebert, and the entire team at the Centre for Biodiversity Genomics for this contribution to transparent benchmarking in biodiversity genomics.
We're bioinformaticians too. spcfy grew out of AIM.science, a DNA metabarcoding lab with over a decade of hands-on experience, and is built by a team that includes practicing bioinformaticians.
spcfy is built for people who run metabarcoding pipelines regularly, not once.
spcfy is your solution when
you process metabarcoding data on a recurring basis, for your own research or for clients, and need results that hold up over time.
spcfy might not be your solution if
you need a one-off analysis of a single dataset, with no requirement to compare it against anything else later.
Some frequently asked questions.
Can I set my own parameters and reference databases?
Yes. Read-length filters and singleton handling are yours to set, along with the choice between OTU clustering (metabarcoding pipeline) and ESV denoising (eDNA pipeline), and your own regional or custom reference databases. BIN assignment for CO1 data runs automatically against BOLD, that part isn't a user setting. spcfy standardizes the infrastructure, not your scientific decisions.
Does spcfy replace or complement my own pipeline?
Most bioinformaticians use spcfy to hand off routine processing and re-runs, while keeping experimental design, QC decisions and interpretation. It replaces the plumbing, not the science.
Who owns my data?
You do. Private workspaces, access rights you define yourself, and export in standard formats at any time. EU hosted, GDPR compliant.
How does onboarding work?
The demo call walks through your workflow and data first, and from there we set up your first project together on the platform.
What does it cost?
Full details are on spcfy.io/pricing — spcfyBASE itself is free, you only pay per processed sample.
What exactly happens in the demo session?
About 30 minutes, with a live walkthrough of the platform on data similar to yours, a conversation about your specific workflow and needs, and, if it's a good fit, the first steps of setting up your own project. No obligation.
What's actually running under the hood?
Paired-end reads are merged with USEARCH, adapters trimmed with Cutadapt, and everything quality-filtered and dereplicated with VSEARCH. From there the two pipelines diverge: the metabarcoding pipeline pre-clusters with VSEARCH and removes chimeras de novo, with final OTU clustering on SWARM v3.1.0 for CO1 amplicons and VSEARCH for other markers, while the eDNA pipeline skips clustering entirely and denoises directly into ESVs with VSEARCH. Any sample that loses more than 30% of its reads between merging and quality filtering pauses automatically so you can exclude it before postprocessing. Dataset-wide singletons are always removed as noise; sample-wide singletons are your call, keep them for sensitivity to rare taxa or drop them for a cleaner table.
Book a demo, or reach out directly
Pick a time on the calendar below for a live walkthrough of the platform, about 30 minutes, no commitment. Prefer to write instead? Send us a short message below.
Tell us about your project
Share a few details about your workflow and your data. Our team usually responds within one business day with a next step that fits.
Standardized, from FASTQ to decision-ready results.
See it on your own data before deciding anything.
Book a live demo ↑