Skip to content

Small Variant Annotation

AIVA integrates a small variant annotation engine to annotate VCF files during upload. Small Variant Annotation adds consequence predictions, gene symbols, transcript information, and links to external databases for every variant in your file.

What Small Variant Annotation Adds

When Small Variant Annotation is applied, the following information is added to each variant and made available as individual columns in the Data Table:

Annotation Description
Consequence The predicted effect of the variant on the transcript (e.g., missense_variant, frameshift_variant, synonymous_variant).
Gene Symbol The HGNC gene symbol associated with the variant.
IMPACT Severity classification: HIGH, MODERATE, LOW, or MODIFIER.
Transcript The transcript ID affected by the variant.
Protein Position The amino acid position affected in the protein product.
Amino Acid Change The reference and alternate amino acids (e.g., R/H).
SIFT Prediction of whether the amino acid change affects protein function (Tolerated / Deleterious) with a confidence score.
PolyPhen Prediction of the structural and functional impact of the amino acid change (Benign / Possibly Damaging / Probably Damaging) with a score.
Existing Variation Known variant identifiers (e.g., rsIDs from dbSNP).

These fields are parsed from the consequence annotation (CSQ) string and broken out into individual, sortable, filterable columns.

Subscription Requirements

Small Variant Annotation is available on the following subscription tiers:

Tier Available
Free No
Trial Yes
Plus Yes
Pro Yes

See Subscription Tiers for a full comparison of plan features.

How to Enable Small Variant Annotation

  1. Navigate to the Samples section and open the upload dialog.
  2. Select or drag-and-drop a VCF file (.vcf or .vcf.gz).
  3. Check the Enable Small Variant Annotation checkbox.
  4. Click Upload to submit the file.

The system creates a multi-step background job:

  1. Upload/Download: The file is uploaded (or downloaded from a cloud URL).
  2. Small Variant Annotation: The annotation engine runs against the VCF file, adding CSQ annotations.
  3. Parsing: The annotated file is parsed, and CSQ subfields are extracted into individual columns.
  4. Ready: The sample appears in your sample list with all annotation columns available.

Monitor progress in the Job Manager.

VCF files only

Small Variant Annotation is available only for VCF files. CSV and TSV files cannot be annotated, as the annotation engine requires the VCF format to interpret variant calls correctly.

Assembly Support

AIVA supports two reference genome assemblies for Small Variant Annotation:

  • GRCh38 (hg38): The current human reference genome assembly. This is the default.
  • GRCh37 (hg19): The previous assembly, still widely used in clinical settings.

The assembly is configured at the server level. Ensure your VCF files are aligned to the same reference assembly configured on your AIVA instance.

Processing Time

Small Variant Annotation processing time depends on the number of variants in your file:

File Size Approximate Time
< 10,000 variants Under 1 minute
10,000 to 100,000 variants 1 to 5 minutes
100,000 to 1,000,000 variants 5 to 30 minutes
> 1,000,000 variants 30 minutes or more

Continue working while annotation runs

Small Variant Annotation runs as a background job. You can navigate away, upload additional files, chat with AIVA, or work on other samples while the annotation completes. You will receive a notification when the job finishes.

Using Annotation Columns in Analysis

Once annotation is complete and the sample is ready:

  • Data Table: Annotation columns appear in the column chooser. Add them to your table view to see consequence predictions, gene symbols, SIFT/PolyPhen scores, and more alongside your variant data. See Navigation and Layout.
  • Filtering: Filter on annotation columns like any other column. For example, filter by Consequence = missense_variant or IMPACT = HIGH. See Filtering.
  • AI Analysis: Ask AIVA questions that reference annotations. For example, "Show me all HIGH impact variants with deleterious SIFT predictions." AIVA queries these columns directly.
  • Export: Annotation columns are included in CSV exports when they are visible in the table.

Server-Side Requirements

Administrator note

Small Variant Annotation requires a local annotation cache on the server. The cache is approximately 18 GB in size and must be downloaded separately. See the setup documentation for installation instructions. The feature will not function if the cache is not present.