AI portrait
This portrait was algorithmically built from this dog's genome: their genotype at 8 morphology loci (coat length, curl, color, ear set, body size, head shape, skull, furnishings) plus their position within the CanVAS Founder atlas. The same dog always reproduces the same portrait. A different dog with different alleles gets a different portrait.
GRLS_grls9GSIT1NN
GRLS_grls9GSIT1NN is a Golden Retriever from the GRLS research cohort. One of 18,477 dogs in the atlas.
See GRLS_grls9GSIT1NN in the atlasGRLS_grls9GSIT1NN is a strong genetic outlier within the Golden Retriever cluster - among the most distinctive examples of their breed.
- Predicted medium by the six body-size genes the atlas reads (IGF1, HMGA2, SMAD2, LCORL, STC2, ADAMTS17).
- Standard leg length. No chondrodysplasia retrogene variant.
- Wire-coat furnishings - the eyebrow + beard variant at RSPO2.
The five dogs in the atlas whose genomes sit closest to GRLS_grls9GSIT1NN's. Click any of them to keep exploring.
GRLS_grls9GSIT1NN sits in the Golden Retriever cluster, with genome overlap to Labrador Retriever - sister breeds nearby in the atlas.
Breed similarity from non-negative least squares against 91 breed centroids in PCA-256 space, corrected for atlas sample-size imbalance. Without correction, Goldens (22% of the atlas) leak into every dog's raw NNLS breakdown; with it, the bias falls out. Raw fractions stay in the dataset for re-derivation. Methodology.
- Golden Retriever 75%
- Labrador Retriever 25%
From the Golden Retriever Lifetime Study . Breed-page reference: Golden Retriever.
The actual allele call at each locus's representative SNP for this dog. Each gene name links to its page where you can see the per-breed frequency table and the direction of effect.
Technical details click to expand
The numbers behind the placement. Useful for researchers reproducing the math or debugging an unexpected position; not interesting to most readers.
y 11.388
z 8.030
The 3 PCs on which GRLS_grls9GSIT1NN scores most extreme, with the 3 highest-loading SNPs on each. Foundation for the future genome-ring visualization.
- chr22:26,809,841 loading 0.0329
- chr13:7,517,248 loading -0.0323
- chr26:11,829,785 loading 0.0322
- chr9:10,719,447 loading 0.0424
- chr34:1,972,619 loading -0.0401
- chr34:1,432,211 loading -0.0400
- chr13:1,243,516 loading 0.0372
- chr13:4,612,002 loading -0.0358
- chr13:2,021,999 loading -0.0342