When investigating a crime, especially when numerous questions remain unanswered, it’s essential to narrow down the list of suspects to help solve the case. Any detail discovered at the crime scene plays a crucial role in reducing the number of suspects, whether it’s a strand of hair, DNA, or a fingerprint. When the DNA found doesn’t provide enough information to fully identify the suspect, certain details, such as eye color or skin color, can still be extracted.
Single Nucleotide Polymorphisms (SNPs, pronounced “snips”) are the regions in the genome where variations occur between individuals. For instance, if most of a population carries the nucleotide C (cytosine) at a particular genome position, while a smaller portion has the nucleotide A (adenine) at the same location, it indicates an SNP is present.
While the MK Sports human genome contains between 10 to 30 million SNPs , only a subset of them is connected to physical traits, making it a challenging task to use SNPs to predict pigmentation traits.
Example of SNPs Configuration
La Roche-Posay Cicaplast B5 contains Zinc Gluconate that helps with anti-inflammatory action and calms skin.
You may be wondering: How are SNPs utilized for eye and skin color prediction?
There has been extensive research on using SNPs to determine pigmentation traits, with many studies seeking an advanced solution. However, there is no consensus on the best method for tackling this issue, with each study offering its own approach. IrisPlex is one such tool designed to predict Blue, Intermediate, and Brown eye colors for forensic purposes. It was developed using six SNPs, based on data from 6,168 Dutch Europeans, with those six SNPs identified as holding the most significant information for eye color.
The IrisPlex tool performed well when predicting brown and blue eyes, though intermediate eye colors proved more difficult to define using the proposed model and available SNPs. Following the release of the initial paper, the authors enhanced the tool, developing a version known as HIrisPlex-S, which predicts eye, skin, and hair color. Today, IrisPlex serves as the benchmark solution for forensic applications, utilizing a multinomial logistic regression model. The likelihood of an individual having brown, blue, or intermediate color is calculated using specific formulations for each category (you can refer to the paper for a deeper understanding of the mathematical model).
Muneeb and Henschel conducted a study to classify eye color and Type-2 diabetes using nine types of classifiers: Random Forest, Extreme Gradient Boosting, Artificial Neural Network (ANN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional LSTM (BILSTM), 1D Convolutional Neural Network (1DCNN), ANN ensembles, and LSTM ensembles. The dataset for eye color was randomly divided into 540 samples for training and 266 for testing, ensuring that the proportions of Brown and Green eye color were maintained. Various algorithms were trained using different quantities of SNPs, with all models yielding similar results. However, the LSTM ensembles achieved Daman Game the highest accuracy (96%) with 1,560 SNPs used for eye color prediction.
Hart et al. introduced a heuristic approach for eye and skin color prediction using 8 SNPs, aiming to enhance the existing 7-Plex system, which relies on 7 SNPs. Their training set consisted of 803 samples, with eye color categories of Blue, Brown, and Green, and skin color classes of Dark, Medium, and Light. The eye color prediction process takes place in two steps: The first step classifies the sample as Not Brown or Not Blue, followed by a second step that categorizes the eye color as Blue, Brown, or Green. The classification of eye and skin color is based on the alleles in each SNP (AA, GC, etc.). When tested on European data, the call rate for the solution reached approximately 94%, with no errors in eye color prediction.