Talk:Quantitative trait

Talk

Discussion for: Quantitative trait • View • Links • Related changes

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

Page scope

Quantitative trait covers what a quantitative trait is, how such traits are recorded and how their variation is divided, what each kind of evidence can say about the number of loci, which cannabis characters have been treated as quantitative, and yield. The variance components and published estimates belong to Heritability, mapping methods and populations to Quantitative trait locus, the interaction with environment to Genotype-by-environment interaction, the response of one genotype to Phenotypic plasticity, the inheritance of chemotype, sex and day-length insensitivity to Genetic traits, and gene action to Dominance and recessivity.

Data verification

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

Discussion

Sources

Hill (2010), Philosophical Transactions of the Royal Society B 365: 73–85: a personal perspective on the statistical model, its assumptions, the infinitesimal model, and the numbers of loci found in other species. It concentrates on animal breeding.
Beavis, Mahama and Suza (2023), "Plant Breeding Basics", and Beavis and colleagues (2023), "Gene Effects", "Components of Variance" and "G x E", in Quantitative Genetics for Plant Breeding: scales of measurement, Fisher's model, the division of variance and the interaction term.
Lübberstedt, Beavis and Suza (2023), "Cluster Analysis, Association, and QTL Mapping" and "Marker Assisted Selection and Genomic Selection", in Molecular Plant Breeding: what mapping can resolve, and marker selection on polygenic traits.
Woods and colleagues (2021), Genetics 219: iyab099: Carmagnola × USO31.
Salentijn, Petit and Trindade (2019), Frontiers in Plant Science 10: 614: a review of flowering behaviour and fibre quality in hemp.
Petit and colleagues (2020), Frontiers in Plant Science 11: 102 and 11: 569958, and Frontiers in Genetics 11: 566314: the panel of 123 hemp accessions.
Carlson and colleagues (2021), Journal of Experimental Botany 72: 7694–7709: 23 hemp families in one field, flowering, form and yield.
Naim-Feil and colleagues (2021), BMC Plant Biology 21: 294, and (2023), Plants 12: 493: medicinal genotypes grown as clones.
de Ronne and colleagues (2024), Scientific Reports 14: 9162; de Ronne and Torkamaneh (2025), The Plant Genome 18: e70031; Yoosefzadeh Najafabadi and Torkamaneh (2024), The Plant Journal 121: e17164: the Canadian drug-type panel.
Backer and colleagues (2019), Frontiers in Plant Science 10: 495: meta-analysis of published cannabis yields.
Hennink (1994), Euphytica 78: 69–76: selection in Dutch fibre hemp.
Grassa and colleagues (2021), New Phytologist 230: 1665–1679; Toth and colleagues (2022), Frontiers in Plant Science 13: 991680; Manansala-Siazon and colleagues (2025), Plants 14: 3853; Seifi and colleagues (2025), Frontiers in Plant Science 16: 1543229: single crosses.
Faux and colleagues (2014), Euphytica 196: 183–197, and (2016), Euphytica 209: 357–376: sex expression.
McDonald and Lubell-Brand (2024), HortScience 59: 1795–1799: selfed lines and F1 hybrids.
Babaei and Torkamaneh (2026), The Plant Genome 19: e70243, and Mostafaei Dehnavi and colleagues (2025), BMC Plant Biology 25: 80: Iranian populations.

Where the sources disagree

Few loci or many. Salentijn and colleagues (2019) state as general that flowering time, sex determination and fibre quality in hemp are governed by many loci. Woods and colleagues (2021) attribute the agronomic and biochemical differences between two hemp cultivars to few regions of large effect, while noting that their models sum to about half the variance. The statements differ in scope: the first concerns the species and the second one cross.
"Qualitative" and "quantitative" for cannabinoids. Naim-Feil and colleagues (2023) call the ratio between cannabinoids a qualitative trait and absolute amounts quantitative. In the sense of a measurement scale, both are quantitative: the ratio in Skunk No. 1 × Carmen was mapped as a measured trait.
Are quantitative characters the plastic ones? Flowering dates vary by degree and in the 123-accession panel had large genetic components with small interactions with site. Height, stem weight and absolute cannabinoid amounts moved with the environment. The sources support the second group as examples and give no general rule that quantitative characters are more plastic than others.
Heritability of floral biomass. Carlson and colleagues (2021) give h² of 0.81 for dry stripped floral biomass, calculated as four times the family variance over the phenotypic variance. Naim-Feil and colleagues (2021) give a broad-sense value of 0.33 for floral bud dry weight in clones. Narrow-sense heritability cannot exceed broad-sense heritability in one population. These are different populations and designs, the first a set of families that share a parent.
SNP-based heritability in the Iranian panel. Babaei and Torkamaneh (2026) report SNP-based h² of 1.00 for total THC potential alongside H² of 0.90, a SNP-based estimate at its ceiling and above the broad-sense one.
"Single QTL" in Woods and colleagues. The discussion says single QTLs were identified for seed yield, thousand-seed mass and dry biomass. Table 1 of the same paper gives five, four and four.
Significant SNPs in Mostafaei Dehnavi and colleagues. The abstract speaks of "approximately 23,266 significant high-quality" SNPs. The results give 23,266 as the filtered marker set used for population structure and association, so the figure does not count associated markers.

How far the evidence goes

Each of the crosses describes two parents or one family set. Carmagnola × USO31 was one field in 2017, the Carlson families one field in 2020.
Most association results come from one site in one season, the drug-type panel from legal Canadian market accessions and the Iranian panels from populations of one country.
A single peak or a skewed distribution is read by de Ronne and colleagues and by Babaei and Torkamaneh as a sign of several loci or of rare alleles. Neither paper tests that reading by itself, and each goes on to map the traits.
In the drug-type panel, rare variants were filtered out at a minor allele frequency of 6 per cent, so it cannot speak to rare alleles.
The sources give no estimate of the number of loci for yield in cannabis. Markers and QTLs have been reported, and their variance shares come from single samples.
Genomic prediction in the cited test covers cannabinoids only. It has no yield trait.
The fresh and dry weights of the flowers in the Iranian panel did not differ significantly among pools in the first analysis. Comparison of the genetic groups in the same paper found a significant difference in dry flower weight.

Terms

QTL: quantitative trait locus.
Polygenic: used by the sources for traits under many genes. Some use "complex trait" in the same sense.
H² and h²: broad-sense and narrow-sense heritability.
Sexual maturity and harvest maturity in de Ronne and colleagues (2024): the first floral bud at the base of an axillary stem, and days to maturity.
Dry stripped floral biomass in Carlson and colleagues (2021): dried floral material stripped by hand, about 60 per cent of dry biomass.

Further figures from the sources

Carlson and colleagues (2021), ranges and h²: height 78 to 255 cm, 0.83; leaflet number 3 to 7, 0.80; pre-terminal flowering day 28 to 85 days after planting, 0.85; powdery mildew severity 1 to 95 per cent, 0.67; dry stripped floral biomass 0.27 to 2.15 kg, 0.81; cannabinoid yield 25 to 320 g, 0.87.
Naim-Feil and colleagues (2021): floral bud dry weight 15 to 210 g per plant; harvest index 10 to 30 per cent.
De Ronne and colleagues (2024), drug-type panel: fresh biomass 90 to 1,260 g; plant height 22 to 109 cm; first flower bud 20 to 68 days.