01 The Plant Genome · 21 September 2026_
A pangenome-based approach reveals genes associated with polyploidy and apomixis in Eragrostis curvula
- Population genetics
A short-read linear pangenome of weeping lovegrass shows polyploid cytotypes carry more non-reference sequence than diploids, and lists candidate genes for polyploidy and apomixis.
Summary
Eragrostis curvula is a forage grass with diploid and polyploid cytotypes that switch between sexual and apomictic reproduction. The authors built a linear pangenome from a diversity panel by iteratively mapping Illumina reads to the reference and assembling the unmapped reads, then used it for gene presence/absence calls, annotation and variant calling. Polyploid accessions contained a larger fraction of the newly assembled sequence than diploids. Genes associated with polyploidy were enriched for cell-cycle and DNA-replication functions; genes associated with apomixis were enriched for post-transcriptional regulation. Allele-frequency distributions and phylogenies also separated cytotypes.
What to take from it
- Iterative map-then-assemble linear pangenome from short reads, not a graph
- Polyploids carry more non-reference sequence than diploids
- Polyploidy-associated genes: cell cycle and DNA replication
- Apomixis-associated genes: post-transcriptional regulation
Review
The method is the older iterative-assembly approach: short-read contigs from unmapped reads are fragmented and collapse paralogs and homeologs, which matters most in exactly the polyploid accessions where the extra sequence was found. Part of the "more genes in polyploids" result may be a detection effect, since higher ploidy gives more distinct haplotypes to assemble from the same data. Ploidy and reproductive mode are confounded in this species, so separating polyploidy genes from apomixis genes needs a design where the two vary independently; the abstract does not say how many accessions break that link. Before using the candidate lists: panel size per cytotype, PAV validation by PCR or long reads, and any correction for population structure.
02 Frontiers in Plant Science · 7 September 2026_
A gene-based pangenome reveals genomic divergence and altitudinal differentiation in allotetraploid quinoa
- Population genetics
Eleven quinoa genomes give a near-closed gene pangenome of 51,298 orthogroups, with presence/absence differences between highland and lowland accessions that the authors themselves flag as inflated by uneven sampling.
Summary
The authors built an orthogroup-based pangenome from 11 quinoa genomes, 7 highland and 4 lowland: 51,298 orthogroups, 19,653 core, with the curve close to saturation. Core genes showed the strongest purifying selection, but the difference from other gene classes was not significant. Subgenome A kept fewer single-copy genes than subgenome B. PAV analysis found 7,477 orthogroups enriched in highland and 3,549 in lowland accessions; highland-enriched groups included transposon-related and FAR1 families, lowland-enriched groups nitrogen metabolism and defence. Copy number of betalain genes did not explain seed coat colour differences. Four positively selected genes carry population-specific amino acid changes.
What to take from it
- 11 genomes, 51,298 orthogroups, 19,653 core
- Subgenome A retains fewer single-copy genes than B
- Highland vs lowland PAV counts are descriptive only (7 vs 4 accessions)
- Betalain gene copy number does not explain seed coat colour
Review
The authors state that the ecotype PAV counts are probably inflated by the 7-versus-4 design, which leaves the altitude-differentiation headline mostly descriptive. "Near-closed" from 11 genomes says little about the species; it reflects the sampled accessions. A gene-based pangenome misses the non-genic structural variation that usually drives adaptation calls. The selection tests on four genes need to be checked against demography. Useful as an orthogroup catalogue for quinoa; for adaptation claims, wait for a graph pangenome with balanced ecotype sampling and population-scale resequencing.
03 Journal of Integrative Plant Biology · 22 September 2026_
Pan-genomic analysis of alternative polyadenylation reveals a 3'UTR remodeling regulatory layer controlling salt tolerance in rice
- Population genetics
Across more than 200 rice accessions, salt stress shortens 3'UTRs, and about 1,800 cis-QTL for polyadenylation site choice map to different regions than expression QTL; two genes were validated by CRISPR.
Summary
The authors profiled alternative polyadenylation (APA) in over 200 rice accessions under normal and salt-stress conditions, finding about 20,000 and 22,000 APA events. Salt stress shifted use toward proximal polyA sites, shortening 3'UTRs, which can remove miRNA binding sites from stress-response genes. They mapped 1,634 and 1,809 cis-3'aQTL under the two conditions; these were enriched in 3'UTRs and termination sites, whereas eQTL were enriched in promoters. CRISPR knockout and 3'UTR deletion of OsMyb21 and OsRNS3 changed salt tolerance.
What to take from it
- Salt stress globally shortens rice 3'UTRs via proximal polyA sites
- 1,634 (normal) and 1,809 (salt) cis-3'aQTL
- 3'aQTL sit in 3'UTRs and termination sites; eQTL in promoters
- OsMyb21 and OsRNS3 validated by knockout and 3'UTR deletion
Review
The "pan-genomic" label refers to a population panel; the abstract does not say whether a pangenome reference was used. If reads went to one reference, APA sites in non-reference sequence and structural variants that remove polyA signals would be missed or miscalled. That is the obvious extension for a pangenome reference. The 3'aQTL/eQTL separation is a useful result, but whether 3'aQTL add phenotypic signal beyond eQTL needs colocalisation with salt-tolerance GWAS, which the abstract does not report. The validation shows the two genes affect tolerance; it shows APA-mediated regulation only if the 3'UTR deletion alone reproduces the phenotype, which is worth checking in the figures.
04 G3: Genes, Genomes, Genetics · 22 September 2026_
Evolutionary dynamics of polymorphic endogenous retrovirus insertions across wild house mouse populations
- Structural variation
- Population genetics
Short-read data from 163 wild mice across Eurasia yield over 100,000 non-reference ERV insertions, with subspecies-specific patterns and adaptive introgression of the antiviral Fv4 insertion into Korean mice.
Summary
Endogenous retroviruses are a major class of structural variation in mammals, but their diversity in wild populations is poorly described. The authors wrote a new pipeline to detect and annotate non-reference ERV insertions from short reads and applied it to 163 wild Mus musculus genomes. They found over 100,000 non-reference insertions, with differences in number, genomic location and population specificity among subspecies and populations. The ERV-derived Fv4 gene, which gives resistance to murine leukaemia viruses, is restricted to certain subspecies; in the Korean population its spread was attributed to adaptive introgression from neighbouring populations.
What to take from it
- 163 wild mouse genomes, over 100,000 non-reference ERV insertions
- Insertion patterns differ by subspecies and population
- Fv4 antiviral ERV spread into Korean mice by adaptive introgression
- New short-read ERV insertion caller
Review
Short-read detection of non-reference insertions finds the junctions but usually not the full inserted sequence, so ERV family assignment and age estimates depend on how the pipeline resolves the insert; precision and recall against long-read assemblies of the same mice are what to look for. Long-read assemblies of a subset of the same wild mice would be the natural truth set. The Fv4 introgression case is the strongest result because it combines a known phenotype with population data; the genome-wide "distinct host-retrovirus histories" interpretation is softer. Directly transferable to livestock: the same approach applies to ERV and other mobile-element polymorphism in cattle and sheep resequencing panels.
05 Animal Genetics · 22 September 2026_
Genetic Diversity and Selection Signatures of Three Indonesian Indigenous Pig Populations From Eastern Indonesia Using Whole Genome Sequencing Data
- Animal science
- Population genetics
Whole genomes of 42 pigs from Maluku, Southeast Sulawesi and East Nusa Tenggara place them with southern Chinese breeds, show European introgression in two populations, and find mostly population-specific sweeps.
Summary
The authors sequenced 42 indigenous pigs from three eastern Indonesian regions and compared them with public Chinese domestic, European domestic and wild boar genomes. The Indonesian pigs cluster with southern Chinese breeds. Maluku pigs are genetically uniform; Southeast Sulawesi and East Nusa Tenggara pigs carry European alleles and have higher diversity. Selective sweep scans gave mostly population-specific signals, with candidate genes including FBXO36 (Maluku), DIAPH3 (East Nusa Tenggara) and GTPBP4 (Southeast Sulawesi).
What to take from it
- 42 new whole genomes from three eastern Indonesian populations
- Closest to southern Chinese breeds
- European introgression in Southeast Sulawesi and East Nusa Tenggara pigs
- Sweeps are mostly population-specific
Review
Useful mainly as new data from an undersampled region. With 42 pigs split across three populations, sweep scans have limited power and high false-positive rates, and the candidate genes are linked to "economic traits" without any phenotype data, so treat them as hypotheses. The missing comparison is with Sus celebensis and other Island Southeast Asian wild Sus; without them, admixture from local wild species could be misread as diversity or selection. The European introgression finding is the most solid result and matters for conservation sampling. Before using: sample sizes per group, the wild Sus references included, and whether sweeps overlap introgressed tracts.