Fenugreek study finds a weak yield link between two Egyptian sites

Fenugreek accessions that produced more seed at one Egyptian field site did not necessarily perform similarly at a second, according to a reanalysis of a plant collection. The researchers also found a genetic marker associated with a measure of productivity across the two environments, but the association does not establish a causal gene.

Maha Aljabri, Mustafa M. H. Abd El-Wahab, Amani Mahmoud Dobeie and Mohamed El-Soda published the study in PLOS One on October 29, 2026. It reexamines yields reported earlier, rather than describing a new 2026 field season. The collection comprised 107 Egyptian fenugreek accessions from 15 governorates, tested at Giza and Wadi El-Natrun with three replicates at each site in one season.

For this panel, the authors report mean seed yields per plant of 102.22 grams at Giza and 148.91 grams at Wadi El-Natrun. The correlation between yields at the two sites was r=0.107, a weak relationship that points to differences in relative performance within the collection. These are measurements per plant under the tested conditions, not yields per hectare or a forecast for farms.

The sites differed in soil, salinity and climate conditions at the same time. Because the study did not impose a single defined stress treatment, it cannot isolate which of those conditions caused the difference or show that salinity improved yields.

The researchers derived nine measures of response, productivity and stability from the same yield records and analyzed 38,142 single-nucleotide polymorphisms, or SNPs—genetic markers that can be compared across the accessions. In a univariate analysis, one marker passed a strict genome-wide threshold for an index the paper calls STI, used here to summarize productivity across the two environments. Separate multivariate results included suggestive signals; the methods reuse the same underlying yields and are not independent replications. Marker coverage was limited, so the scan may have missed some associations. The suggestive findings may also include false positives, and estimates of effects in the multivariate analysis were unstable; those leads need further testing.

Comparative sequence analysis classified the strongest statistical signal as repetitive sequence rather than identifying a causal coding gene. A repetitive sequence can still mark an association, but the finding does not establish which gene or mechanism affects yield. The authors call for work in independent populations, more environments and seasons, and functional validation. The paper reports no specific funding and no competing interests.

Rowan Finch analysis: A breeder faces three different questions: which accessions yielded well in these fields, whether their performance holds elsewhere, and what biological mechanism might explain a pattern. This study offers leads for further testing, but its two sites and one season cannot answer all three. New field comparisons and tests of what a candidate sequence actually does would address different gaps before a practical breeding claim can be made.

This is an independently written explanation and analysis of the authors’ paper, which is available under the Creative Commons Attribution 4.0 licence.