To date, using traditional STR methods, DNA has been used only as a biometric identifier, as though it contained no more information than a fingerprint. However, an individual's DNA contains an abundance of descriptive information — in fact, it contains the genetic blueprint that differentiates one person from another.The Parabon Snapshot™ DNA Phenotyping System takes advantage of modern SNP technology, translating select biomarkers taken from a DNA sample into predictions about various physical traits of its source — a descriptive profile that contains eye color, hair color, skin color, height, handedness, etc., even face morphology.
The Snapshot DNA Phenotyping Service
Soon you will be able to send us a human DNA sample and, using Snapshot, we will produce a forensic profile containing predictions about the physical traits of its source.
Because some traits are partially determined by environmental factors and not DNA alone, Snapshot trait predictions are presented with a corresponding measure of confidence, which reflects the degree to which such factors influence each particular trait. Traits, such as eye color, that are highly heritable (i.e., are not greatly affected by environmental factors) are predicted with higher accuracy and confidence than those that have lower heritability.
The first set of traits supported by Snapshot will include:
- blood type
- hair morphology
- shoe size
- vision correction
- ear lobe attachment
- ear lobe crease
- 5th digit clinodactyly
- flat footedness
- eye color, hair color, skin color
- height, handedness, etc.
Parabon is outfitting Snapshot with a unique ability to predict face morphology and coupling it with other related trait predictions, giving Snapshot the ability to produce a "digital mugshot" of an individual corresponding to a given DNA sample. Although Snapshot provides a single predicted digital mugshot for a given DNA sample, the composite image below illustrates how various genetic influences affect the appearance of faces.
How Snapshot Works
Recent advances in DNA sequencing technology have made it practical and affordable to read genetic content from DNA, which in turn has allowed the creation of datasets that include both genotypic (genetic content) and phenotypic (trait) data for each of thousands of subjects. With the diligent and repeated application of data mining and machine learning processes to such data, Parabon NanoLabs produces statistical models that translate the presence of specific genetic biomarkers into forensically relevant trait predictions.
Beginning with large datasets comprised of a phenotype (trait) of interest and genotype data for thousands of subjects, our bioinformatics team performs large-scale statistical analysis on millions of individual SNPs and billions of combinations thereof to identify sets of these genetic markers that associate with the given trait. This mining process can take weeks running on hundreds, sometimes thousands, of computers. In the end, those SNPs with the greatest likelihood of contributing to the variation observed in the target trait are culled for potential use in predictive models.
The modeling phase further refines this set of SNPs to a final set that most accurately predict the target trait under a framework of machine learning algorithms. Models are validated against data held out for such testing and calibrated with all available data before being installed into the Snapshot architecture.Snapshot will also provide predictions about the kinship relation between the sources of two or more DNA samples. Snapshot mining and modeling is an ongoing process designed to constantly improve the accuracy and confidence of the predictive models that result.