Key terms
Single Nucleotide Polymorphism — a single-letter variation in DNA. Genotyping arrays read thousands to millions of SNPs per animal, the raw material for genomic prediction and parentage.
The set of genetic markers (typically SNPs) recorded for an individual animal. In Helical, genotypes are stored once and re-used for any number of analyses.
An observed or measured trait — weight, growth, milk yield, carcase quality, methane, fertility. Phenotypes plus genotypes and pedigree feed genetic evaluation.
The recorded ancestry of an animal — sire, dam and earlier generations. Pedigree links relatives so information flows across the population during evaluation.
Estimated Breeding Value — the genetic merit of an animal for a trait, expressed as the value it is expected to pass to progeny. EPD (Expected Progeny Difference) is the same idea expressed per offspring.
Best Linear Unbiased Prediction — the statistical method behind modern EBVs, separating genetic merit from environmental effects across all recorded relatives.
Single-step genomic BLUP combines pedigree, phenotype and genomic information in one evaluation — no separate blending step, and no loss of information.
Using genome-wide marker information to predict breeding values, so young animals can be selected accurately before they have their own performance records.
The set of genotyped, phenotyped animals used to estimate marker effects. A larger, well-structured reference population increases the accuracy of genomic predictions.
The proportion of variation in a trait that is due to genetics. It sets how much of an animal’s performance is expected to be inherited by its progeny.
A weighted combination of EBVs across multiple traits into a single number aligned to a breeding objective, so animals can be ranked on overall economic merit.
Confirming an animal’s recorded sire and dam against its genotype. Helical runs this provider-agnostically across labs and chip types.
A heritable trait or disorder governed by specific markers. Helical interprets carrier status with configurable, breed-specific rules.
The probability that the two copies of a gene in an animal are identical by descent — a measure of how related its parents are, used to manage genetic diversity.
The high-performance evaluation engine Helical integrates with for single-step genomic evaluations, using accelerated solving (PCG) and Bayesian methods (MCMC).
Markov Chain Monte Carlo — a sampling technique used in Bayesian evaluation to estimate variance components and marker effects for complex genetic architectures.
Preconditioned Conjugate Gradient — an iterative solver that makes large genomic evaluations fast, enabling more frequent runs on millions of animals.
Filling in untyped markers by inference from a reference panel, so animals genotyped on lower-density panels can be analysed alongside high-density data.
Automated checks on incoming genotypes and phenotypes — call rates, conflicts, outliers — that catch errors before they propagate into evaluations.
Genetic evaluation across breed groups, composites and crossbred populations, modelling genetic heterogeneity and breed-specific effects in one analysis.