How VAMS ATLAS™ works.
A clear view of how sequencing-derived information becomes explainable, layered intelligence — with visible confidence language and traceable structure.
Supported inputs
What ATLAS can read
Availability of any given output reflects the sequencing route and report type you choose.
| 16S rRNA amplicon | Community-level composition and diversity — foundation-level reading. |
|---|---|
| Shotgun metagenomics | Species-level structure and functional-potential context. |
| Metatranscriptomics | Expression-context signals where sample quality supports it — a planned layer at VAMS BIOME. |
| Existing sequencing data | Processed tables or FASTQ can be discussed, subject to QC and method fit. |
Quality & eligibility checks
What is checked before anything is read
From reads to comparable features
Quality-checked sequencing data is turned into reproducible, comparable features — feature tables, taxonomy and diversity summaries — using versioned, documented pipelines so results stay comparable across runs and cohorts.
- ✓Reproducible feature and taxonomy tables
- ✓Diversity summaries appropriate to the method
- ✓Versioned pipelines keep analyses comparable
- ✓Reference context supports interpretation
How a signal becomes a readable score
Each microbiome signal area is summarised into a comparative, plain-language score paired with a confidence cue. The internal weighting and computation are proprietary and are not exposed publicly; what is shown is the readable meaning and its confidence, not the formula.
- ✓Comparative, plain-language scores
- ✓Every score carries a confidence cue
- ✓Internal weighting kept private
- ✓Scores are the readable entry point — not raw abundance
Panel organisation
How related scores are grouped
Evidence & confidence gates
Guardrails are part of how the system works
Each reading is paired with confidence framing and method-aware limits so interpretation stays grounded and proportionate.
How the workflow becomes a readable report
Scores, panels and packs are assembled into a structured report — headline structure down to supporting context — with confidence cues and clear, non-diagnostic framing. The sample report shows how the workflow is translated into a user-facing experience.
- ✓Structured, scannable layout
- ✓Confidence cues shown in context
- ✓Supporting context and definitions
- ✓Illustrative sample data only — not a personal result
Interpretation boundaries
What the workflow does not claim
See the workflow become a report
The sample report shows how sequencing input becomes readable scores, panels and packs — with confidence cues and non-diagnostic framing.
The interpretation step, illustrated
From unordered reads to readable structure.
Sequencing produces a dense, unordered signal. ATLAS™ passes that signal through one interpretation layer and returns it as ordered, readable structure — the step this page walks through.
Sequencing signal
Unordered, method-dependent reads
ATLAS™ interpretation layer
Structured output
Ordered, labelled, bounded by method
Where the structure lands
Each step feeds a layered architecture.
Interpreted composition becomes conservative scores, related scores group into panels, and panels assemble into themed Insight Packs where the chosen method supports them. Each layer states its method support.
Explore the score architecture →- 01 Composition Sequencing-derived microbial features from the sample.
- 02 Scores Conservative summaries of those features, each stating its method support.
- 03 Panels Related scores grouped under a biological theme — the unit of interpretation.
- 04 Insight Packs Panels grouped into a themed report layer, where the method supports it.