Jump to content
Toggle search
Toggle menu
  • 5K articles
  • 1.8K files
  • 88 users
  • 44.5K edits
Landrace.wiki
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Population structure

From Landrace.wiki
More actions
Part of a series on
Genetics
Four rows of vertical bars, one bar for each plant, with sample labels beneath. The top row is green and red. Each row below adds a colour, blue, then purple, then orange, and more of the bars become a single colour.
The same 52 samples assigned by ADMIXTURE to two groups (top row) and up to five (bottom row). Boxes mark the five groups Chen and colleagues report.
Cannabis genetics
Mendelian Genetics

Mendelian inheritance·Dominance and recessivity·Monogenic trait·Dihybrid cross·Epistasis

Sex Determination

Sex determination in cannabis·Sex expression·Hermaphroditism

Quantitative Genetics

Quantitative trait·Heritability·Genotype-by-environment interaction·Response to selection·Genetic correlation·Heterosis

Population Genetics

Genetic diversity·Heterozygosity·Effective population size·Hardy–Weinberg equilibrium·Genetic drift·Gene flow·Population structure·Inbreeding·Inbreeding depression·Introgression·Population bottleneck

Molecular Genetics

Cannabis genome·Genetic marker·Quantitative trait locus·Phylogenetics of Cannabis

Cannabinoid Genetics

Chemotype·THCA synthase·CBDA synthase

Breeding Systems

Outcrossing·Open pollination·Landrace breeding·Farmer selection·F1 hybrid·Recurrent selection

Selection & Adaptation

Natural selection·Artificial selection·Mass selection·Local adaptation·Selective sweep

Genetic Conservation

Conservation genetics·Ex situ conservation·In situ conservation·Genetic erosion

Genetic Threats

Genetic contamination·Hybridisation·Genetic monitoring

Portal·Category

Population structure is the arrangement of genetic variation within and among the populations of a species.[1] Plants that disperse only a short way leave nearby plants more alike than distant ones.[2] Genetic drift makes separated groups diverge and gene flow draws them back together.[3][4]

Studies read structure from samples of plants typed with genetic markers. FST gives the degree of differentiation between groups. An analysis of molecular variance divides the variation among levels such as regions, populations and plants. Principal component analysis plots plants by their similarity and clustering programs assign them to a chosen number of groups. Each result describes the plants sampled. The number of groups can change with the sample and the method.[5][6]

In Cannabis, Jason Sawler and colleagues typed 43 hemp and 81 drug-type samples at 14,031 SNPs and found a mean FST of 0.156 between the two.[7] Anna Halpin-McCormick and colleagues analysed nine collections by one method and found three to seven genetic clusters, with inconsistent structure by use type.[8] Whether any such group is a species is a separate dispute, treated under Cannabis taxonomy.

Why populations come to differ

Populations cut off from one another drift apart. Take one large population divided into subpopulations of N plants with no migration between them. After t generations FST equals 1 − (1 − 1/(2N))t, the inbreeding accumulated by drift: about 0.10 after ten generations at 50 plants.[3] Migration works against this. In the island model every population exchanges migrants with every other. In the stepping-stone model migrants come from neighbours.[9] Limited dispersal of this kind produces isolation by distance, in which plants close together are more alike than plants far apart.[2]

Natural selection adds differences at the genes it acts on, and such regions can stand out in scans of FST along the genome.[10] In a pangenome of 193 cannabis genomes, one SNP in GIGANTEA, a gene of the circadian clock and of flowering time, reached an FST of 0.77 between drug-type plants and hemp.[11]

Across 307 studies of wild plants, long-lived, outcrossing and late-successional species held most of their variation within populations. Annual, selfing or early-successional species held most of it among populations.[12] Cannabis is wind-pollinated and mostly dioecious, which makes it an outcrosser.[13]

Breeding also divides a crop. With 13 microsatellites, Christophe Dufresnes and colleagues found an FST of about 0.39 among 15 drug-type varieties against about 0.15 among 24 hemp varieties. Several of the drug-type varieties consisted of clones of one genotype.[14] Genetic diversity sets out how those values relate to diversity within varieties.

A sample that pools plants from several populations holds more homozygotes than its allele frequencies predict, even if each population mates at random. This is the Wahlund effect, described under heterozygosity. It vanishes once the mixture has mated at random for a generation.[15]

Reading structure from genes

Each method answers a different question about the same genotypes.

Methods for reading population structure
Method Question Result What limits it
FST How far do groups differ in allele frequency?[3] One value for a pair or set of groups Groups are set in advance. The value depends on diversity within populations,[16] on how values are combined and on which SNPs are used[17]
Analysis of molecular variance At which level of a hierarchy does the variation lie?[18] A share of variance for each level Levels are those the study defines
Principal component analysis How are plants arranged along the main axes of variation?[19] A plot of plants, with no groups assigned Sampling affects the axes. Waves in spatial data can be artefacts[6][20]
Model-based clustering (STRUCTURE, ADMIXTURE) What share of each plant's genotype fits each of K groups?[21][22] A bar for each plant, split among K colours K is a choice. The model looks for groups at Hardy–Weinberg and linkage equilibrium[5]
Regression on distance Does differentiation rise with distance?[23] A slope Needs populations of known location
Population graphs (TreeMix) Which tree with migration events fits the allele frequencies?[24] A tree with added migration edges The number of edges is tested and chosen.[11] Several histories can fit the same data[24]

Differentiation

FST goes back to Sewall Wright and Gustave Malécot. Its relation to FIS and FIT is set out under heterozygosity. Published estimates for the same pairs of human populations moved by more than a factor of two in some cases between genotyping arrays and sequence data. Most of the gap came from how the values at single SNPs were combined and from which SNPs were included. The choice of estimator at a single SNP mattered little in that comparison.[17]

The value also depends on diversity within populations. Where gene diversity is high, GST and its relatives approach zero even if the subpopulations are completely differentiated.[25] Patrick Meirmans and Philip Hedrick advise reporting FST with a version standardised by its maximum (F′ST) or with Jost's D.[16] FST is also often converted into a number of migrants per generation, a conversion Michael Whitlock and David McCauley judged rarely accurate, because real populations break the assumptions of the model. They still called FST an excellent measure of differentiation.[26] The uses of that number are taken up under gene flow.

Plots and clusters

Three tree diagrams side by side above three bar plots. Two of the trees have arrows of mixing and the third has one narrow branch. The bar plots beneath look almost the same, each with one group shown as a mix of blue, green and pink.
Three simulated histories of populations P1 to P4: recent admixture, admixture with an unsampled population and a recent bottleneck. ADMIXTURE gives almost the same bars for all three. From Lawson and colleagues (2018).

Luigi Cavalli-Sforza and colleagues first used principal component analysis on genetic data. Below a threshold of divergence that depends on the size of the dataset, structure is essentially undetectable. A little above it, detection is easy.[19] The expected position of a sample on the axes can be predicted from average coalescence times between pairs of genomes, which for two populations ties distance on the plot to FST. Histories that give the same coalescence times therefore cannot be told apart, and uneven sampling strongly affects the projection.[27] Gradients and waves can appear in such plots whenever variation is continuous in space, so they do not necessarily mark particular migrations.[20]

Model-based clustering assumes K populations, each with its own allele frequencies. It assigns each plant to one of them, or jointly to several if its genotype looks admixed.[21] ADMIXTURE uses the likelihood model of STRUCTURE and runs considerably faster.[22] The user tests a range of K, and the methods for choosing among them differ. The ΔK statistic picks out the uppermost level of a hierarchy.[28] In a review of 1,264 studies, 54 per cent of those using ΔK reported K = 2 against 21 per cent of the others. Simulations extending that work often returned two even where more subpopulations existed.[29]

Uneven sampling changes the answer. Thinly sampled subpopulations tend to merge while heavily sampled ones split, even where the heavily sampled plants belong to one randomly mating population.[5] In simulations with too few clusters allowed, all the individuals of the largest sample sometimes fell into one cluster, though Steven Kalinowski found that the program gave appropriate clusters in many circumstances.[30] A bar plot also does not show a history by itself. Daniel Lawson and colleagues simulated a recent admixture, an admixture with an unsampled population and a recent bottleneck. ADMIXTURE gave almost identical plots for all three.[6]

Clusters and gradients

Whether variation falls into clusters or runs in gradients has been argued over in human data. David Serre and Svante Pääbo found gradients of allele frequency when individuals were sampled evenly across the world.[31] Noah Rosenberg and colleagues, with 993 markers, accepted that gradients exist but found that geographic dispersion had comparatively little effect on clustering. They proposed that clusters can arise from small jumps in genetic distance across geographic barriers.[32] In simulated populations on a landscape, STRUCTURE correctly found one population when genotypes were random. Where neighbour mating had built gradients, it found several clusters, whose number changed with the sampling scheme.[33] A gradient sampled only at its two ends can look like two discrete clusters.[19] Isolation by distance can also produce false positives in tests of hierarchical structure.[2] The regression of FST/(1 − FST) on distance, or on its logarithm, estimates population density multiplied by the spread of dispersal.[23]

What structure does to other analyses

An association study can link a marker to a trait because both differ between groups. Structure of this kind has limited association mapping in plants and humans. A mixed model that accounts for relatedness reduced false signals in 277 maize inbred lines.[34] Principal components can correct for the same effect.[35] Models of past population size are affected too: a structured population can give a false signal of change in size, described under population bottleneck.[36]

In cannabis

Published surveys differ in their plants, their markers and the way they choose the number of groups.

Surveys of population structure in cannabis
Study Plants Markers How groups were found Groups reported
Hillig 2005 157 accessions 17 allozyme loci Principal components of 52 alleles Two major groups and a putative third[37]
Gilmore and colleagues 2007 188 plants, 76 populations Seven organelle polymorphisms Parsimony and distance trees Three haplotype groups, "not strictly diagnostic" of use or origin[38]
Sawler and colleagues 2015 43 hemp, 81 drug-type 14,031 SNPs Principal components; fastSTRUCTURE run at K = 2 to test for two groups Hemp and drug type[7]
Lynch and colleagues 2016 195 samples 2,894 SNPs for the tree FLOCK; fastSTRUCTURE FLOCK: three (hemp, narrow-leaflet and broad-leaflet drug types). fastSTRUCTURE: two, broad-leaflet drug types against the rest[39]
Soorni and colleagues 2017 95 plants, 68 of them from Iran and Afghanistan 24,710 SNPs Principal components; discriminant analysis; fineSTRUCTURE Two by principal components and 19 by fineSTRUCTURE among the Iranian plants[40]
Dufresnes and colleagues 2017 1,324 plants of 48 accessions 13 microsatellites STRUCTURE with ΔK Two: hemp and drug type[14]
Zhang Q. and colleagues 2018 43 Chinese populations Five chloroplast regions Spatial analysis of molecular variance Three, by latitude[41]
Henry and colleagues 2020 420 North American samples 23 selected SNPs Discriminant analysis of principal components Five[42]
Zhang J. and colleagues 2020 199 accessions 72 microsatellite loci STRUCTURE with ΔK Two: 99 and 100 accessions, the second all from China[43]
Schwabe and colleagues 2021 49 samples Ten microsatellites STRUCTURE with ΔK; MavericK Hemp apart from drug type, with no clear division of retail Sativa, Hybrid and Indica[44]
Ren and colleagues 2021 104 accessions 12,010,905 SNPs Maximum-likelihood tree; admixture; principal components Four: basal, hemp, feral drug type, drug type[13]
Chen and colleagues 2022 52 samples, 47 of them Chinese Whole-genome SNPs ADMIXTURE at K = 2 to 5; tree; principal components Five groups within China, by source and ecotype[45]
Woods and colleagues 2023 190 samples 8,474,449 SNPs ADMIXTURE, by cross-validation Six[46]
Mostafaei Dehnavi and colleagues 2025 228 plants of 35 Iranian populations About 23,266 SNPs ADMIXTURE, by cross-entropy Five[47]
Aina and colleagues 2025 346 feral plants, of 760 sequenced 21,037 SNPs ADMIXTURE; cross-validation unresolved, Bayesian information criterion used Five[48]
Lynch and colleagues 2025 193 genomes of 144 plants k-mers; phased SNPs Hierarchical k-mer clustering; TreeMix European hemp, Asian hemp and two drug-type groups[11]

Hemp and drug-type samples

A scatter plot of points in four colours on two axes. Blue hemp points lie across the upper left and red points along the right, with orange Iranian points between them, nearer the red. Labelled outliers lie at the lower left, and a blue point labelled CAN 37 97 sits among the orange points.
Iranian plants, genebank accessions and the hemp and drug-type samples of Sawler and colleagues at 13,325 SNPs, on the first two principal components. The hemp accession CAN 37/97 lies on the drug-type side. From Soorni and colleagues (2017).

Sawler and colleagues separated 43 hemp from 81 drug-type samples on the first principal component at 14,031 SNPs. No SNP was fixed for different alleles in the two, and the highest FST at a single SNP was 0.87. They concluded that the two groups share most of one pool of variation.[7] Christopher Grassa and colleagues report an FST of 0.229 between the two categories at 2,051 SNPs in 367 accessions.[49] Lynch and colleagues report 0.20 from phased SNPs in the 193 genomes of their pangenome.[11] Patrick Woods and colleagues, with 190 samples at about 8.5 million SNPs, found 0.116 for their most divergent pair of groups, feral plants of the United States against drug-type plants.[46]

Hemp and drug type do not always form the first division. Ryan Lynch and colleagues found K = 2 most likely with fastSTRUCTURE in 195 samples, but the split separated broad-leaflet drug types from narrow-leaflet drug types and hemp. Some hemp and narrow-leaflet plants were assigned almost wholly to one population, despite their separation on the tree. FLOCK applied to the same samples supported at least three groups.[39] Some plants also fall outside their labelled group. The hemp accession CAN 37/97 clustered with drug-type plants in the survey by Sawler and colleagues, who could not exclude a sample mix-up or a wrong label.[7] Dufresnes and colleagues count it among the limits of their own database that each reference sample's plant type came from its supplier.[14]

Sativa and indica labels

In the survey by Sawler and colleagues, reported sativa ancestry explained about a third of the position of 81 drug-type samples on the first principal component (r2 = 0.36).[7] In 137 samples at 116,296 SNPs, Sophie Watts and colleagues found no clear clustering by label, and the first two components each had an R2 of 0.12 with a scale from Sativa to Indica.[50] Anna Schwabe and Mitchell McGlaughlin typed 122 samples of 30 strains at ten microsatellites and found two genetic groups that did not correspond to the labels. They note that their samples held too few landrace strains to test for such a division among landraces.[51] In a later set of 49 samples at the same loci, Schwabe and colleagues found no clear distinction among retail Sativa, Hybrid and Indica samples, while hemp samples clustered together.[44]

Studies disagree on which label lies nearer hemp. Sawler and colleagues found a lower FST between hemp and strains reported as wholly indica (0.136) than between hemp and strains reported as wholly sativa (0.161). In the microsatellite survey of Dufresnes and colleagues, most of the drug varieties closest to hemp were of sativa ancestry.[7][14] Both took ancestry from the reports of sellers or suppliers. The same words have also meant something else: Karl Hillig used sativa and indica for gene pools drawn from allozyme frequencies, with the indica pool including fibre and seed landraces of eastern Asia, and proposed three species from them.[37] The labels are treated under Sativa vs Indica.

Geography

A map of China with pie charts at sampling sites, mostly blue in the north, red at middle latitudes and green in the south-west. Beside it a network joins numbered haplotypes in the same three colours.
Chloroplast haplogroups of cannabis in China: H (blue) lies mainly at high latitudes, M (red) at middle latitudes and L (green) at low latitudes. Pie charts give the frequencies in each population. From Zhang Q. and colleagues (2018).

With five chloroplast regions, Qingying Zhang and colleagues divided 43 Chinese populations into three groups by latitude. Variance between the groups was 69.48 per cent of the total and FST among them 0.695. The same populations, grouped by wild or domesticated status, showed no significant difference (FST = 0.023).[41] The authors contrast the first result with nuclear studies, in which most variation lay within cultivars, and consider maternal inheritance of the chloroplast the probable reason.[41]

Nuclear surveys within China give other divisions. Jiangjiang Zhang and colleagues, with 199 accessions at 72 microsatellite loci, found a mean FST of 0.0468. STRUCTURE returned two subgroups of about equal size, one of them all Chinese and the other mainly from abroad.[43] Pan and colleagues used 14 insertion and deletion markers to divide 115 Chinese accessions into a group of 84, mainly from the north of China, and a group of 31, mostly from the south. A separate analysis of flowering time also split them between north and south.[52] Chen and colleagues, with whole-genome SNPs in 52 samples, reported five Chinese groups by geographic source and ecotype.[45] In the pangenome of Lynch and colleagues, hemp from Europe and from Asia formed separate groups. Their k-mer clustering also set a single wild Tibetan assembly apart and kept the samples Ren and colleagues called basal and feral drug type as distinct populations.[11]

Landraces and feral populations

The same plants can give different counts. Soorni and colleagues typed one female and one male plant from each of 35 Iranian locations. Principal components gave two clusters and fineSTRUCTURE 19, some of them matching single locations.[40] Mostafaei Dehnavi and colleagues later reported five groups in 228 plants from 35 Iranian populations, with about 23,266 SNPs.[47] In Morocco, six microsatellites set the local population that Chaymae Benkirane and colleagues call Beldia apart from the introduced cultivar Critical.[53] Lynch and colleagues cautioned that strains sold as landraces may be admixed, which can be a problem for STRUCTURE and fastSTRUCTURE.[39]

Ademola Aina and colleagues sequenced 760 feral plants from 12 states of the United States. Cross-validation could not settle the number of clusters, so they used the Bayesian information criterion, which fell until K = 5. They kept the 346 plants with more than 70 per cent membership in one group, which removes admixed plants from the analyses that follow. FST among the five groups ran from 0.07 to 0.39, lowest among three groups in the Midwest and highest for New York against Indiana.[48]

The share of variance assigned to each level differs among studies, and so do the levels.

Division of molecular variance in six cannabis studies
Study Markers Levels and shares
Zhang Q. and colleagues 2018 Chloroplast regions 69.48 per cent between three groups, 14.43 among populations within them, 16.10 within populations[41]
Piluzza and colleagues 2013 DNA polymorphisms 74 per cent among accessions, 26 within[54]
Aina and colleagues 2025 21,037 SNPs About 9 per cent between groups, 20 between accessions within groups, 71 among all samples[48]
Benkirane and colleagues 2024 Six microsatellites 65 per cent within individuals, 19 within populations, 16 among populations[53]
Soorni and colleagues 2017, two clusters SNPs 93.09 per cent within populations, 1.38 between[40]
Soorni and colleagues 2017, 19 clusters SNPs 95.74 per cent within populations, 1.02 between[40]

Aina and colleagues suggest that such differences may come from sample size, sample composition and marker type.[48]

How far the studies can be compared

Halpin-McCormick and colleagues ran one method, hierarchical clustering of principal components, across nine independent collections. From the three to seven clusters they found, they stressed the importance of sampling where passport data are limited.[8] Counts from different studies are not estimates of one quantity. Sawler and colleagues fixed K at two to test for two groups, Dufresnes and colleagues and Zhang J. and colleagues used ΔK, and Woods and colleagues and Mostafaei Dehnavi and colleagues took the lowest cross-validation or cross-entropy error. ΔK itself leans towards two.[7][14][43][46][47][29]

The groups compared are different kinds of thing: hemp and drug-type categories, commercial varieties, genebank accessions, regions of origin, field locations, and clusters that the method itself has defined. The markers also differ. Chloroplast markers follow the maternal line.[41] Microsatellite loci are usually chosen for high polymorphism,[55] and high diversity holds FST down.[25] SNP sets are filtered before analysis, at a minor allele frequency of 1 per cent in one survey and 5 per cent in another.[7][46] Values of 0.156, 0.20 and 0.229 between hemp and drug type come from different plants, SNP sets and calculations. Calculation alone can shift such a value: in human data, combining the same SNPs another way moved one estimate from 0.106 to 0.072.[17]

Agreement between surveys is not always independent. The 367 accessions of Grassa and colleagues draw on the data of Sawler, Lynch and Soorni and their colleagues. Soorni and colleagues reanalysed the Sawler samples. Woods and colleagues took their drug-type samples from Lynch and colleagues.[49][40][46] Access to samples was complicated by the illegal status of the drug and by the few hemp repositories in international seedbanks.[7]

See also

References

  1. ↑ Davinack, D. (2024). Molecular Ecology & Evolution: An Introduction. Norton, Massachusetts: Wheaton College. p. 6.
  2. ↑ 2.0 2.1 2.2 Meirmans, P.G. (2012). "The trouble with isolation by distance". Molecular Ecology. 21 (12): 2839–2846. doi:10.1111/j.1365-294x.2012.05578.x.
  3. ↑ 3.0 3.1 3.2 Felsenstein, J. (December 2019). Theoretical evolutionary genetics. Seattle: Department of Genome Sciences, University of Washington. pp. 275–277.
  4. ↑ Ellstrand, N.C. (2014). "Is gene flow the most important evolutionary force in plants?". American Journal of Botany. 101 (5): 737–753. doi:10.3732/ajb.1400024.
  5. ↑ 5.0 5.1 5.2 Puechmaille, S.J. (2016). "The program STRUCTURE does not reliably recover the correct population structure when sampling is uneven: subsampling and new estimators alleviate the problem". Molecular Ecology Resources. 16 (3): 608–627. doi:10.1111/1755-0998.12512.
  6. ↑ 6.0 6.1 6.2 Lawson, D.J.; van Dorp, L.; Falush, D. (2018). "A tutorial on how not to over-interpret STRUCTURE and ADMIXTURE bar plots". Nature Communications. 9 (1) 3258. doi:10.1038/s41467-018-05257-7.
  7. ↑ 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 Sawler, Jason; Stout, Jake M.; Gardner, Kyle M.; Hudson, Darryl; Vidmar, John; Butler, Laura; Page, Jonathan E.; Myles, Sean (2015). "The genetic structure of marijuana and hemp". PLOS ONE. 10 (8) e0133292. doi:10.1371/journal.pone.0133292.
  8. ↑ 8.0 8.1 Halpin-McCormick, Anna; Heyduk, Karolina; Kantar, Michael B.; Batora, Nicholas L.; Masalia, Rishi R.; Law, Kerin B.; Kuntz, Eleanor J. (2024). "Examining population structure across multiple collections of Cannabis". Genetic Resources and Crop Evolution. 71 (8): 4705–4722. doi:10.1007/s10722-024-01928-1.
  9. ↑ Felsenstein, J. (December 2019). Theoretical evolutionary genetics. Seattle: Department of Genome Sciences, University of Washington. pp. 180–181.
  10. ↑ Holsinger, K.E.; Weir, B.S. (2009). "Genetics in geographically structured populations: defining, estimating and interpreting FST". Nature Reviews Genetics. 10 (9): 639–650. doi:10.1038/nrg2611.
  11. ↑ 11.0 11.1 11.2 11.3 11.4 Lynch, R.C.; Padgitt-Cobb, L.K.; Garfinkel, A.R.; Knaus, B.J.; Hartwick, N.T.; et al. (2025). "Domesticated cannabinoid synthases amid a wild mosaic cannabis pangenome". Nature. 643 (8073): 1001–1010. doi:10.1038/s41586-025-09065-0. PMC 12286863.
  12. ↑ Nybom, H. (2004). "Comparison of different nuclear DNA markers for estimating intraspecific genetic diversity in plants". Molecular Ecology. 13 (5): 1143–1155. doi:10.1111/j.1365-294x.2004.02141.x.
  13. ↑ 13.0 13.1 Ren, G.; Zhang, X.; Li, Y.; Ridout, K.; Serrano-Serrano, M.L.; Yang, Y.; Liu, A.; Ravikanth, G.; Nawaz, M.A.; Mumtaz, A.S.; Salamin, N.; Fumagalli, L. (2021). "Large-scale whole-genome resequencing unravels the domestication history of Cannabis sativa". Science Advances. 7 (29) eabg2286. doi:10.1126/sciadv.abg2286.
  14. ↑ 14.0 14.1 14.2 14.3 14.4 Dufresnes, C.; Jan, C.; Bienert, F.; Goudet, J.; Fumagalli, L. (2017). "Broad-scale genetic diversity of Cannabis for forensic applications". PLoS ONE. 12 (1) e0170522. doi:10.1371/journal.pone.0170522.
  15. ↑ Felsenstein, J. (December 2019). Theoretical evolutionary genetics. Seattle: Department of Genome Sciences, University of Washington. pp. 175–176.
  16. ↑ 16.0 16.1 Meirmans, P.G.; Hedrick, P.W. (2011). "Assessing population structure: FST and related measures". Molecular Ecology Resources. 11 (1): 5–18. doi:10.1111/j.1755-0998.2010.02927.x.
  17. ↑ 17.0 17.1 17.2 Bhatia, G.; Patterson, N.; Sankararaman, S.; Price, A.L. (2013). "Estimating and interpreting FST: the impact of rare variants". Genome Research. 23 (9): 1514–1521. doi:10.1101/gr.154831.113.
  18. ↑ Excoffier, L.; Smouse, P.E.; Quattro, J.M. (1992). "Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data". Genetics. 131 (2): 479–491. doi:10.1093/genetics/131.2.479.
  19. ↑ 19.0 19.1 19.2 Patterson, N.; Price, A.L.; Reich, D. (2006). "Population structure and eigenanalysis". PLOS Genetics. 2 (12) e190. doi:10.1371/journal.pgen.0020190.
  20. ↑ 20.0 20.1 Novembre, J.; Stephens, M. (2008). "Interpreting principal component analyses of spatial population genetic variation". Nature Genetics. 40 (5): 646–649. doi:10.1038/ng.139.
  21. ↑ 21.0 21.1 Pritchard, J.K.; Stephens, M.; Donnelly, P. (2000). "Inference of population structure using multilocus genotype data". Genetics. 155 (2): 945–959. doi:10.1093/genetics/155.2.945.
  22. ↑ 22.0 22.1 Alexander, D.H.; Novembre, J.; Lange, K. (2009). "Fast model-based estimation of ancestry in unrelated individuals". Genome Research. 19 (9): 1655–1664. doi:10.1101/gr.094052.109.
  23. ↑ 23.0 23.1 Rousset, F. (1997). "Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance". Genetics. 145 (4): 1219–1228. doi:10.1093/genetics/145.4.1219.
  24. ↑ 24.0 24.1 Pickrell, J.K.; Pritchard, J.K. (2012). "Inference of population splits and mixtures from genome-wide allele frequency data". PLOS Genetics. 8 (11) e1002967. doi:10.1371/journal.pgen.1002967.
  25. ↑ 25.0 25.1 Jost, L. (2008). "GST and its relatives do not measure differentiation". Molecular Ecology. 17 (18): 4015–4026. doi:10.1111/j.1365-294x.2008.03887.x.
  26. ↑ Whitlock, M.C.; McCauley, D.E. (1999). "Indirect measures of gene flow and migration: FST ≠ 1/(4Nm + 1)". Heredity. 82 (2): 117–125. doi:10.1038/sj.hdy.6884960.
  27. ↑ McVean, G. (2009). "A genealogical interpretation of principal components analysis". PLOS Genetics. 5 (10) e1000686. doi:10.1371/journal.pgen.1000686.
  28. ↑ Evanno, G.; Regnaut, S.; Goudet, J. (2005). "Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study". Molecular Ecology. 14 (8): 2611–2620. doi:10.1111/j.1365-294x.2005.02553.x.
  29. ↑ 29.0 29.1 Janes, J.K.; Miller, J.M.; Dupuis, J.R.; Malenfant, R.M.; Gorrell, J.C.; Cullingham, C.I.; Andrew, R.L. (2017). "The K = 2 conundrum". Molecular Ecology. 26 (14): 3594–3602. doi:10.1111/mec.14187.
  30. ↑ Kalinowski, S.T. (2011). "The computer program STRUCTURE does not reliably identify the main genetic clusters within species: simulations and implications for human population structure". Heredity. 106 (4): 625–632. doi:10.1038/hdy.2010.95.
  31. ↑ Serre, D.; Pääbo, S. (2004). "Evidence for gradients of human genetic diversity within and among continents". Genome Research. 14 (9): 1679–1685. doi:10.1101/gr.2529604.
  32. ↑ Rosenberg, N.A.; Mahajan, S.; Ramachandran, S.; Zhao, C.; Pritchard, J.K.; Feldman, M.W. (2005). "Clines, clusters, and the effect of study design on the inference of human population structure". PLOS Genetics. 1 (6) e70. doi:10.1371/journal.pgen.0010070.
  33. ↑ Schwartz, M.K.; McKelvey, K.S. (2009). "Why sampling scheme matters: the effect of sampling scheme on landscape genetic results". Conservation Genetics. 10 (2): 441–452. doi:10.1007/s10592-008-9622-1.
  34. ↑ Yu, J.; Pressoir, G.; Briggs, W.H.; Vroh Bi, I.; Yamasaki, M.; Doebley, J.F.; et al. (2006). "A unified mixed-model method for association mapping that accounts for multiple levels of relatedness". Nature Genetics. 38 (2): 203–208. doi:10.1038/ng1702.
  35. ↑ Price, A.L.; Patterson, N.J.; Plenge, R.M.; Weinblatt, M.E.; Shadick, N.A.; Reich, D. (2006). "Principal components analysis corrects for stratification in genome-wide association studies". Nature Genetics. 38 (8): 904–909. doi:10.1038/ng1847.
  36. ↑ Mazet, O.; Rodríguez, W.; Grusea, S.; Boitard, S.; Chikhi, L. (2016). "On the importance of being structured: instantaneous coalescence rates and human evolution: lessons for ancestral population size inference?". Heredity. 116 (4): 362–371. doi:10.1038/hdy.2015.104.
  37. ↑ 37.0 37.1 Hillig, K.W. (2005). "Genetic evidence for speciation in Cannabis (Cannabaceae)". Genetic Resources and Crop Evolution. 52 (2): 161–180. doi:10.1007/s10722-003-4452-y.
  38. ↑ Gilmore, S.; Peakall, R.; Robertson, J. (2007). "Organelle DNA haplotypes reflect crop-use characteristics and geographic origins of Cannabis sativa". Forensic Science International. 172 (2–3): 179–190. doi:10.1016/j.forsciint.2006.10.025.
  39. ↑ 39.0 39.1 39.2 Lynch, R.C.; Vergara, D.; Tittes, S.; White, K.; Schwartz, C.J.; Gibbs, M.J.; et al. (2016). "Genomic and chemical diversity in Cannabis". Critical Reviews in Plant Sciences. 35 (5–6): 349–363. doi:10.1080/07352689.2016.1265363.
  40. ↑ 40.0 40.1 40.2 40.3 40.4 Soorni, A.; Fatahi, R.; Haak, D.C.; Salami, S.A.; Bombarely, A. (2017). "Assessment of genetic diversity and population structure in Iranian Cannabis germplasm". Scientific Reports. 7 (1) 15668. doi:10.1038/s41598-017-15816-5.
  41. ↑ 41.0 41.1 41.2 41.3 41.4 Zhang, Q.; Chen, X.; Guo, H.; Trindade, L.M.; Salentijn, E.M.J.; Guo, R.; Guo, M.; Xu, Y.; Yang, M. (2018). "Latitudinal adaptation and genetic insights into the origins of Cannabis sativa L." Frontiers in Plant Science. 9 1876. doi:10.3389/fpls.2018.01876.
  42. ↑ Henry, P.; Khatodia, S.; Kapoor, K.; Gonzales, B.; Middleton, A.; Hong, K.; et al. (2020). "A single nucleotide polymorphism assay sheds light on the extent and distribution of genetic diversity, population structure and functional basis of key traits in cultivated north American cannabis". Journal of Cannabis Research. 2 (1) 26. doi:10.1186/s42238-020-00036-y.
  43. ↑ 43.0 43.1 43.2 Zhang, J.; Yan, J.; Huang, S.; Pan, G.; Chang, L.; Li, J.; et al. (2020). "Genetic diversity and population structure of Cannabis based on the genome-wide development of simple sequence repeat markers". Frontiers in Genetics. 11 958. doi:10.3389/fgene.2020.00958.
  44. ↑ 44.0 44.1 Schwabe, Anna L.; Hansen, Connor J.; Hyslop, Richard M.; McGlaughlin, Mitchell E. (29 September 2021). "Comparative Genetic Structure of Cannabis sativa Including Federally Produced, Wild Collected, and Cultivated Samples". Frontiers in Plant Science. 12 675770. doi:10.3389/fpls.2021.675770. PMC 8544287.
  45. ↑ 45.0 45.1 Chen, Xuan; Guo, Hong-Yan; Zhang, Qing-Ying; Wang, Lu; Guo, Rong; Zhan, Yi-Xun; Lv, Pin; Xu, Yan-Ping; Guo, Meng-Bi; Zhang, Yuan; Zhang, Kun; Liu, Yan-Hu; Yang, Ming (2022). "Whole-genome resequencing of wild and cultivated cannabis reveals the genetic structure and adaptive selection of important traits". BMC Plant Biology. 22 (1) 371. doi:10.1186/s12870-022-03744-0.
  46. ↑ 46.0 46.1 46.2 46.3 46.4 Woods, P.; Price, N.; Matthews, P.; McKay, J.K. (2023). "Genome-wide polymorphism and genic selection in feral and domesticated lineages of Cannabis sativa". G3. 13 (2) jkac209. doi:10.1093/g3journal/jkac209.
  47. ↑ 47.0 47.1 47.2 Mostafaei Dehnavi, M.; Damerum, A.; Taheri, S.; Ebadi, A.; Panahi, S.; et al. (2025). "Population genomics of a natural Cannabis sativa L. collection from Iran identifies novel genetic loci for flowering time, morphology, sex and chemotyping". BMC Plant Biology. 25 (1) 80. doi:10.1186/s12870-025-06045-4. PMC 11748290.
  48. ↑ 48.0 48.1 48.2 48.3 Aina, Ademola; Wenger, Jonathan P.; Stanton, Eliot; Majumdar, Chandrani Gon; ElSohly, Mahmoud; Weiblen, George D.; Ellison, Shelby (1 July 2025). "Genetic diversity, population structure, and cannabinoid variation in feral Cannabis sativa germplasm from the United States". Scientific Reports. 15 (1) 20423. doi:10.1038/s41598-025-07912-8. PMC 12215616.
  49. ↑ 49.0 49.1 Grassa, C.J.; Weiblen, G.D.; Wenger, J.P.; Dabney, C.; et al. (2021). "A new Cannabis genome assembly associates elevated cannabidiol (CBD) with hemp introgressed into marijuana". New Phytologist. 230 (4): 1665–1679. doi:10.1111/nph.17243. PMC 8248131.
  50. ↑ Watts, S.; McElroy, M.; Migicovsky, Z.; Maassen, H.; van Velzen, R.; Myles, S. (2021). "Cannabis labelling is associated with genetic variation in terpene synthase genes". Nature Plants. 7 (10): 1330–1334. doi:10.1038/s41477-021-01003-y.
  51. ↑ Schwabe, Anna L.; McGlaughlin, Mitchell E. (2019). "Genetic tools weed out misconceptions of strain reliability in Cannabis sativa: implications for a budding industry". Journal of Cannabis Research. 1 (1) 3. doi:10.1186/s42238-019-0001-1.
  52. ↑ Pan, G.; Li, Z.; Huang, S.; Tao, J.; Shi, Y.; Chen, A.; et al. (2021). "Genome-wide development of insertion-deletion (InDel) markers for Cannabis and its uses in genetic structure analysis of Chinese germplasm and sex-linked marker identification". BMC Genomics. 22 (1) 595. doi:10.1186/s12864-021-07883-w.
  53. ↑ 53.0 53.1 Benkirane, C.; Charif, M.; Müller, C.M.; Taaifi, Y.; Mansouri, F.; et al. (2024). "Population structure and genetic diversity of Moroccan cannabis (Cannabis sativa L.) germplasm through simple sequence repeat (SSR) analysis". Genetic Resources and Crop Evolution. 71 (5): 2037–2051. doi:10.1007/s10722-023-01754-x.
  54. ↑ Piluzza, G.; Delogu, G.; Cabras, A.; Marceddu, S.; Bullitta, S. (2013). "Differentiation between fiber and drug types of hemp (Cannabis sativa L.) from a collection of wild and domesticated accessions". Genetic Resources and Crop Evolution. 60 (8): 2331–2342. doi:10.1007/s10722-013-0001-5.
  55. ↑ Putman, A.I.; Carbone, I. (2014). "Challenges in analysis and interpretation of microsatellite data for population genetic studies". Ecology and Evolution. 4 (22): 4399–4428. doi:10.1002/ece3.1305.
Retrieved from "https://landrace.wiki/index.php?title=Population_structure&oldid=45436"