Export Formats

Export guide RNA analysis results in CSV, TSV, JSON, or FASTA format for downstream use.

Overview

SPACER supports three export formats, each suited to different downstream workflows. All formats include the full set of scored and ranked guides from your analysis, along with their metadata and quality annotations.

Format	Best For	Includes Metadata
CSV	Spreadsheet analysis, lab notebooks, bulk ordering	Yes
TSV	Bioinformatics pipelines, command-line tools (cut, awk)	Yes
JSON	Programmatic consumption, API integration, data pipelines	Yes
FASTA	Sequence databases, alignment tools, oligo synthesis	Partial (header only)

CSV

Comma-separated values with a header row. Each row represents one candidate guide RNA. All scoring components, quality flags, tier classification, and positional information are included as columns.

csv

rank,spacer_sequence,position,strand,score,tier,gc_content,homopolymer_penalty,activity_score,flags
1,AUGCUUACGAUCGAUCGAU,142,+,92.4,Excellent,0.50,0.0,0.88,
2,GAUCGAUCGUAGCUAGCUA,87,+,85.1,Excellent,0.55,0.0,0.79,
3,CUAGCUAGCUAUCGAUCGA,203,-,71.3,Good,0.45,0.05,0.62,HIGH_GC
4,AAAUGCUUACGAUCGAAAA,310,+,38.7,Poor,0.35,0.15,0.41,HOMOPOLYMER;LOW_ACTIVITY

JSON

Structured JSON output containing the full analysis result including job metadata, pipeline configuration, and an array of guide objects. This is the most complete export format and mirrors the API response structure.

json

{
  "metadata": {
    "sequence_length": 1200,
    "enzyme": "cas13",
    "pam": null,
    "spacer_length": 28,
    "elapsed_ms": 842,
    "ml_active": true,
    "ml_model": "adapt",
    "ml_error": null,
    "structure_active": false,
    "structure_error": null
  },
  "guides": [
    {
      "identifier": "spacer_001_142_164_fwd",
      "sequence": "AUGCUUACGAUCGAUCGAUAGCU",
      "pam": null,
      "crna": null,
      "start": 142,
      "end": 164,
      "orientation": "+",
      "spacer_score": 0.924,
      "activity_score": 0.88,
      "model_name": "adapt",
      "tier": "Excellent",
      "score_breakdown": { "base_score": 70.0, "gc_adjustment": 5.2, "..." : "..." },
      "quality_flags": { "gc_content": 0.50, "gc_optimal": true, "..." : "..." },
      "secondary_structure": null
    }
  ],
  "statistics": { "total_count": 84, "tier_counts": { "excellent": 12, "..." : "..." }, "..." : "..." },
  "primers": null
}

FASTA

Standard FASTA format with guide metadata encoded in the header line. Each entry contains the spacer sequence. Headers include rank, position, score, and tier for quick reference.

text

>guide_1 pos=142 strand=+ score=92.4 tier=Excellent
AUGCUUACGAUCGAUCGAU
>guide_2 pos=87 strand=+ score=85.1 tier=Excellent
GAUCGAUCGUAGCUAGCUA
>guide_3 pos=203 strand=- score=71.3 tier=Good
CUAGCUAGCUAUCGAUCGA

Tip

Use FASTA export when submitting guides directly to oligo synthesis vendors. Most vendors accept FASTA format for bulk ordering.

Export Filtering

By default, all candidate guides are included in the export. You can filter the export to include only specific tiers or guides above a minimum score threshold. Filtering is applied at export time and does not modify the underlying analysis results.

By tier: Export only Excellent and Good guides for a focused set
By score: Set a minimum assay score threshold
By selection: Manually select individual guides from the results table

Info

The export endpoint (POST /v1/export) accepts the same filtering parameters. See the API Reference for details.

FASTA Auto-Format

FASTA exports automatically use the appropriate nucleic acid alphabet based on enzyme family. Cas12 guides are exported as DNA sequences (A/T/G/C) since Cas12 targets double-stranded DNA, while Cas13 guides are exported as RNA sequences (A/U/G/C) reflecting Cas13's single-stranded RNA targeting mechanism.

Variant Analysis Exports

When variant analysis is enabled, additional export data is available for each spacer's cross-variant performance. The variant export includes one row per variant per spacer with the following fields:

Field	Description
Variant ID	Identifier from the input FASTA header
Frequency	Number of input sequences represented by this variant
Activity	Predicted activity score against this variant
Signal ratio	Variant activity divided by on-target mean activity
Signal class	Discrete classification (e.g., high signal, low signal)
Mismatches	Number of mismatches relative to the reference spacer
Is covered	Whether the variant meets the coverage activity threshold
Target sequence	The target subsequence at the spacer position

Aggregate coverage statistics — total variants, covered variants, and coverage percentage — are included alongside the per-variant rows.

Optimizer Result Exports

BADGERS optimizer results export per-site data including optimized spacer sequences with full scoring metadata. Each site includes:

Optimized spacers — sorted by fitness (descending), with composite scores, tier assignments, and quality flags
Consensus fitness — baseline fitness of the natural consensus spacer at each site
Site diversity — number of unique target sequences, Shannon entropy, and valid sequence count
Partition metrics (variant-id mode) — mean on-target and off-target activity scores for evaluating variant discrimination