An information-theoretic approach to estimating the composite genetic effects contributing to variation among generation means: Moving beyond the joint-scaling test for line cross analysis

Summary

Ingested 2026-04-21. 1 findings extracted and verified.

Findings worth citing

Finding 1 — Across 22 re-analyzed empirical LCA datasets, the information-theoretic SAGA approach identified 11 epistatic composite genetic effects (in 9 datasets) with high variable importance (vi > 0.5) that were missed by the traditional joint-scaling test.

For example, nine of the datasets had one or more epistatic CGE (11 in total) not identified with the J-S test that had a vi > 0.5 in the I-T analysis. — p. 429

Why this is citable: This quantifies a central empirical result of the paper — that SAGA’s information-theoretic approach identifies epistatic composite genetic effects missed by the traditional joint-scaling test — and is directly citable for anyone evaluating the prevalence of epistasis in LCA studies or the relative power of I-T vs. hypothesis-testing model selection.

Counter / limitation: The 22 datasets are a non-random convenience sample drawn heavily from the authors’ own prior work (15 of 22 from Demuth & Wade 2007a/b on Tribolium castaneum, 5 from Demuth et al. 2014 on Silene, and 2 from Miller et al. 2003 on D. mojavensis), so the apparent rate at which J-S misses epistasis may reflect the specific cross designs and taxa studied rather than a general bias; additionally, the vi > 0.5 threshold is an author-chosen cutoff with no formal error-rate control, so the count of 11 ‘missed’ epistatic effects is not directly comparable to results from significance-based methods.

Topics: epistasis, line_cross_analysis

Read the paper

doi.org/10.1111/evo.12844

Other papers by these authors

Question copied. Paste it into the NotebookLM tab.