> For the complete documentation index, see [llms.txt](https://docs.nationalism.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nationalism.io/ancestral-group-verification-process/genetic-testing/genetic-analysis-approaches.md).

# Genetic analysis approaches

Once you have the raw genetic data from the testing methods like autosomal SNP genotyping, WGS, WES, STR testing, Y-DNA testing or mtDNA testing, you can then apply a wide range of genetic analysis methods. These methods are the statistical, computational and comparative methods used to interpret that data for ancestry, lineage, kinship and population history.

## **Quality control and preprocessing methods**

These are not ancestry analyses by themselves, but they are necessary before reliable ancestry analysis can happen.

* **1. Sample quality control -** Checks whether a DNA sample is usable and whether genotype/sequence quality is high enough. Helps to prevent false ancestry signals caused by bad data, contamination, low coverage or technical errors.
* **2. Variant calling -** Convert raw sequencing reads into identified genetic variants. Helps to create the list of SNPs/indels used for downstream ancestry and lineage analysis.
* **3. Genotype calling -** Assign genotypes at measured SNP sites from chip or sequence data. Helps to produce the AA/AG/GG-style marker data needed for population comparison.
* **4. Phasing -** Estimates which variants are on the same chromosome copy. This is essential for haplotype analysis, chromosome painting, local ancestry and better IBD detection.
* **5. Imputation -** Predict untyped variants using reference panels. Helps to Increase marker density, improving population analyses and fine-scale ancestry inference.
* **6. Relatedness filtering / sample deduplication -** Identify duplicate or closely related samples in datasets. Helps to prevent family clusters from biasing population analyses.
* **7. Ancestry-informative marker selection -** Choose markers that best distinguish populations. Helps to improve efficiency and sometimes clarity of ancestry classification.

## **Comparing people with relatives**

* **Kinship analysis** - Estimates the degree of biological relationship between individuals. Helps with showing whether people are siblings, cousins, or otherwise related, supporting recent ancestral connections.
* **IBD detection** - Finds long DNA segments shared because of recent common ancestry. Helps provide strong evidence of recent biological descent or overlap with a group through shared ancestors.
* **Parentage testing** - Tests whether one individual is the biological parent of another. Helps with confirming direct recent descent and anchors genealogical lineages.
* **Pedigree methods -** Uses genetic relationship data across multiple people to test or reconstruct family-tree structures. Helps with validating genealogical hypotheses, linking multiple relatives into one family network, and checking whether a proposed pedigree is genetically consistent.

## **Tracing one paternal or maternal line**

* **Y-STR matching -** Compares STR markers on the Y chromosome between men. Helps to support paternal-line connections and lineage matching.
* **mtDNA sequence matching -** Compares mitochondrial sequences across individuals. Helps to support maternal-line connections.
* **Y haplogroup assignment -** Places a male Y chromosome into a branching paternal lineage tree. Helps to identify deep paternal ancestry and relationships among paternal lines.
* **mtDNA haplogroup assignment -** Places mitochondrial DNA into a maternal lineage tree. Helps to Identify deep maternal ancestry and maternal-line relationships.
* **Haplogroup phylogeny -** Uses mutations to place lineages on a branching evolutionary tree. Helps to show how paternal or maternal lineages relate to one another and to broader ancestral branches.
* **TMRCA estimation -** Estimates time to the most recent common ancestor for a lineage. Helps to find a date when two Y or mtDNA lineages shared a common ancestor.

## **Compare someone to a population**

* **Principal component analysis (PCA) -** Compresses genome-wide variation into a few axes of genetic similarity. Helps to show which populations an individual clusters near genetically.
* **Multidimensional scaling (MDS) -** Another way to visualise genetic similarity or distance. Similar to PCA, it helps show population structure.
* **Allele-frequency population matching -** Compares an individual’s allele profile with population frequency profiles. Estimates which populations the person most resembles.
* **Genetic distance analysis -** Uses statistical measures of genetic difference or similarity between individuals or populations. Helps identify which populations are genetically closest and quantify relative separation.
* **Fst analysis -** A population differentiation measure. Helps to quantify how distinct two populations are genetically.
* **Cluster analysis -** Groups individuals into data-driven genetic clusters based on overall similarity. Helps reveal population substructure and possible ancestral groupings without requiring known family relationships.
* **Admixture / model-based clustering** — Estimates ancestry proportions by fitting an individual’s genome as a mixture of reference or inferred population clusters. Helps identify mixed ancestry and population structure.
* **Supervised ancestry assignment -** Assigns ancestry using predefined reference groups. Helps when testing fit to known candidate ancestral populations.
* **Admixture dating -** Estimates when different ancestral populations mixed. Helps determine whether shared ancestry is recent or older.

## **Detailed segment-level ancestry**

These methods use linked DNA segments across the genome and are often more powerful than broad marker-frequency methods.

* **Haplotype-based ancestry inference -** Compares long linked blocks of DNA to reference populations. Helps with distinguishing nearby or subtly different populations.
* **Chromosome painting -** “Paints” each chromosome as copied from various reference populations. Helps with showing which parts of the genome resemble which ancestral groups.
* **Local ancestry inference -** Assigns each genome segment to likely ancestral origins. Helps with showing segment-by-segment ancestry rather than one overall percentage.
* **FineSTRUCTURE-like clustering -** Clusters individuals using haplotype-sharing patterns. Helps with detecting fine-scale subpopulation structure.
* **Haplotype sharing analysis -** Measures how much long-range haplotype material is shared between people or groups. Helps with identifying close population relationships and founder groups.
* **Rare-variant sharing analysis -** Compares uncommon variants that may be population-specific. Helps with strengthening fine-scale ancestry inference and recent population relatedness.

## **Direct historical population comparisons**

* **f-statistics (f3, f4) -** Formal tests of shared genetic drift and admixture. Helps with testing whether a population shares ancestry with another or is admixed from multiple sources.
* **D-statistics -** Tests asymmetry in allele sharing among populations. Helps determine whether one population is genetically closer to one reference than another.
* **qpAdm-like modeling -** Models a target population as a mixture of specified source populations. Helps with testing historical ancestry models.
* **qpGraph-like modeling -** Builds population graphs with splits and admixture edges. Helps reconstruct more complex population histories.
* **Treemix-like analysis -** Tree-based modelling with possible migration events. Helps visualise branching and gene flow among populations.
* **Direct ancient-modern affinity analysis -** Measures genetic similarity or shared ancestry between modern individuals/populations and ancient DNA samples. Helps test whether modern people show affinity to historical populations.
* **Ancient ancestry proportion modelling -** Estimates how much ancestry in a modern population can be explained by ancient source groups. Helps connect present-day populations to historically earlier populations.
* **Temporal population continuity analysis -** Tests whether genetic data are consistent with continuity between populations from different time periods, rather than replacement or major admixture. Helps assess whether a modern group plausibly descends from a historical one.

## **Understanding historical demographic background**

These look at how groups formed, mixed, expanded or remained isolated. \*\*\*\*

* **Runs of homozygosity (ROH) analysis -** Detects long stretches where both chromosome copies are identical. Helps with suggesting endogamy, bottlenecks, founder effects or recent relatedness within populations.
* **Endogamy analysis -** Uses patterns such as excess relatedness, homozygosity, and haplotype sharing to test for repeated within-group marriage. Helps explain why some populations are genetically distinctive.
* **Founder-effect analysis -** Uses allele frequencies, haplotypes, and relatedness patterns to detect descent from a small founding group. Helps identify historically isolated or bottlenecked populations.
* **Effective population size inference -** Estimates historical breeding population size from genetic data. Helps reconstruct demographic history of ancestral populations.
* **Bottleneck analysis -** Uses changes in diversity, homozygosity, and demographic modeling to detect past sharp reductions in population size. Helps explain drift and present-day distinctiveness.
* **Migration/gene-flow analysis -** Estimates movement of genes between populations. Helps determine how populations mixed over time.
* **Coalescent modelling -** Statistical modelling of genealogical history backward through time. Used to infer demographic and ancestral scenarios.