The March of the Beetles: Epistatic Components Dominate Divergence in Dispersal Tendency in Tribolium castaneum

Summary

Ingested 2026-04-21. 2 findings extracted and verified.

Findings worth citing

Finding 1 — Artificial selection on dispersal in T. castaneum produced a rapid response in the first three generations (P2 rising from 25% to 59% dispersal; P1 dropping to 5%) but little further change by generation five.

The base population had a mean dispersal tendency of 25%. Over 3 generations of selection, the dispersal in the P2 line increased to 59% while dispersal tendency in the P1 line reduced to 5%. However, continued selection led to little change; the means in the fifth generation were 70% and 18% for the P2 and P1 lines respectively. — p. 501

Why this is citable: Provides quantitative data on the rate and limits of artificial selection response for dispersal behavior in T. castaneum, including evidence of early rapid response followed by a plateau, relevant for citations about additive genetic variance in behavioral/life-history traits in beetles.

Counter / limitation: Only 3 replicates per selection direction were used from a single source population; generation 4 data are absent due to a confounding procedural error (delayed phenotyping), so the apparent plateau between generations 3 and 5 rests on only two data points and one transitional generation is uninterpretable.

Topics: dispersal, artificial_selection, quantitative_genetics, Tribolium

Finding 2 — Forward-time simulations show that allelic dispersion among parental lines can produce spurious inferences of epistasis in line cross analysis, but only at magnitudes far smaller (epistatic:additive ratio 0–0.33) than observed empirically (ratio 5.27).

For the empirical dataset, this value is 5.27. In contrast, for simulated datasets, the ratio ranged from zero to 0.33. This indicates that dispersion alone is unlikely to explain the large contribution of epistatic component that we infer from our empirical data. — p. 503

Why this is citable: Provides a simulation-based control demonstrating that allelic dispersion in short-term artificial selection lines can produce spurious LCA inferences of epistasis, but that the magnitude of such artifacts (epistatic:additive ratio 0–0.33) is far smaller than the empirically observed ratio (5.27), supporting the conclusion that the inferred epistasis in T. castaneum dispersal is not merely an artifact of incomplete allele fixation.

Counter / limitation: The simulation uses only 20 unlinked biallelic loci with all dispersal alleles dominant and a specific allele-frequency architecture matching the empirical lines; different numbers of loci, linkage structures, dominance relationships, or starting allele frequencies could produce higher false-positive epistatic magnitudes and may not fully bracket the range of possible dispersion artifacts.

Topics: epistasis, line_cross_analysis, dispersal, quantitative_genetics, Tribolium

Read the paper

doi.org/10.1093/jhered/esaa030

Assessing the opportunity for selection to impact morphologi… — 2 shared topics (epistasis, line_cross_analysis)
Wright was right: leveraging old data and new methods to ill… — 2 shared topics (epistasis, line_cross_analysis)
Blackmon & Demuth 2016 — 2 shared topics (epistasis, line_cross_analysis)