Talk:Heterozygosity

Talk

Discussion for: Heterozygosity • View • Links • Related changes

Use this page for verification notes, disputes, and action items. Keep narrative content on the article.

Page scope

Heterozygosity covers heterozygosity as a concept: heterozygotes and homozygotes at a single locus, expected heterozygosity under Hardy–Weinberg proportions, the departures measured by F_IS (inbreeding, the Wahlund effect, the excess in clones of hybrids), how heterozygosity is measured with markers and whole genomes, and the published estimates for cannabis. The measures of variation within and among populations in general belong to Genetic diversity. The loss of heterozygosity by chance in small populations belongs to Genetic drift, its loss under selfing and the cannabis breeding system to Outcrossing, and the fitness consequences of inbreeding to Inbreeding depression.

Data verification

Record source conflicts, figures that differ between studies and sources not used here. Keep narrative content on the article.

Discussion

Sources

Felsenstein (2019), Theoretical Evolutionary Genetics: the Hardy–Weinberg proportions and their history (pp. 4–8); rare alleles in heterozygotes and multiple alleles (pp. 10–11); overdominance and protected polymorphism (pp. 88–91); the Wahlund effect (pp. 175–176); the inbreeding coefficient and genotype frequencies under inbreeding (p. 219); selfing (pp. 231–233); crossing inbred lines (pp. 242–244); the loss of heterozygosity by drift (pp. 254–256); hierarchical F statistics (pp. 275–277); drift and mutation in the infinite-alleles model (pp. 284–285, 289–290).
Davinack (2024), Molecular Ecology & Evolution: An Introduction: expected heterozygosity as gene diversity (pp. 110–112); inbreeding depression (pp. 172–174).
van Bakel and colleagues (2011), Genome Biology 12: R102: heterozygosity of the Purple Kush, Chemdawg, Finola and USO-31 genomes; heterozygous regions in the draft assembly.
Sawler and colleagues (2015), PLoS ONE 10: e0133292: 43 hemp and 81 drug-type samples at 14,031 SNPs; heterozygosity higher in hemp; the method of counting heterozygosity per plant and its expected error.
Dufresnes and colleagues (2017), PLoS ONE 12: e0170522: 1,324 plants of 24 hemp and 15 drug-type varieties at 13 microsatellite loci; observed heterozygosity, F_IS and allelic richness per variety; clonal drug-type breeds with an excess of heterozygotes.
Soorni and colleagues (2017), Scientific Reports 7: 15668: heterozygosity of 68 plants from Iran and Afghanistan; earlier estimates in hemp and drug-type cannabis.
Barcaccia and colleagues (2020), Frontiers in Plant Science 11: 573299: outcrossing as the natural mating system; inbred lines, F₁ hybrids and heterosis in cannabis breeding.
Grassa and colleagues (2021), New Phytologist 230: 1665–1679: the Skunk No. 1 × Carmen mapping population; heterozygosity of the CBDRx, FL18, FL48 and FL49 genomes from k-mer counts; FL48 heterozygous at the THCA synthase locus.
Aina and colleagues (2025), Scientific Reports 15: 20423: observed and expected heterozygosity and F_IS in five clusters of feral plants of the United States; B locus genotypes of about 1,400 feral plants.
Lynch and colleagues (2025), Nature 643: 1001–1010: SNP heterozygosity of the pangenome assemblies; heterozygosity including structural variants; the inversion on chromosome 1; Wright's F_IS as computed in the study.

Where the sources disagree

Whole-genome heterozygosity. The four genomes of van Bakel and colleagues (2011) gave 0.18–0.26 per cent, from short reads aligned to the Purple Kush draft. The four cultivars of Grassa and colleagues (2021) gave 0.38–0.58 per cent from k-mer counts. Lynch and colleagues (2025) give 1–2.5 per cent SNP heterozygosity for their assemblies and "greater than 2%" in their introduction, and 20.6 per cent of genome length once structural variants and regions that cannot be aligned are counted. The studies used different plants and different methods.
Hemp against drug-type cannabis. Sawler and colleagues (2015), Dufresnes and colleagues (2017) and most of the studies Soorni and colleagues (2017) review found hemp more heterozygous. Soorni and colleagues cite one study of 195 samples and 2,894 SNPs (Lynch and colleagues 2016, Critical Reviews in Plant Sciences 35: 349–363) that found the reverse.

Figures that are not comparable

The heterozygosity of a plant in genotyping-by-sequencing studies (0.15 in Soorni and colleagues) is a share of the SNPs genotyped. The whole-genome values (0.18–0.58 per cent) are shares of all bases. Observed heterozygosity at microsatellites (0.41–0.59 in Dufresnes and colleagues) is measured at loci with several alleles each. The three kinds of value are not on one scale.

Terms

Observed heterozygosity (H_O): the share of plants in a sample, or of sites in one plant, that are heterozygous.
Expected heterozygosity (H_E), also gene diversity: the share expected from the allele frequencies under Hardy–Weinberg proportions; for two alleles 2p(1 − p).
F_IS: (H_E − H_O)/H_E; positive for a deficit of heterozygotes and negative for an excess.
Inbreeding coefficient (f): the probability that the two gene copies at a locus in a plant are identical by descent.
Allelic richness: the mean number of alleles per locus, scaled to a common sample size (eight plants in Dufresnes and colleagues).

Images reserved in the article

Two images have places in the article but are not yet uploaded:

File:Cannabis-heterozygosity-Hardy-Weinberg-proportions.jpg (lead), a diagram to be drawn: the Hardy–Weinberg proportions p², 2p(1 − p) and (1 − p)² against the allele frequency p from 0 to 1, the heterozygote arch emphasised and marked at its peak of one half at p = 0.5. No data points, species or allele names.
File:Cannabis-heterozygosity-B-locus-genotypes-feral-United-States.jpg (Heterozygotes at a single locus), a chart to be drawn: three bars for the genotypes at the B locus among feral plants of the United States genotyped by Aina and colleagues (2025), B₁B₁ 1,120, B₁B₂ 208 and B₂B₂ 91, each labelled with the chemotype the model of inheritance assigns to it (III, II and I). No expected values or percentages.