Author: slquinlan

Am J Health Syst Pharm. 2026 Sep 10:zxag282. doi: 10.1093/ajhp/zxag282. Online ahead of print.

ABSTRACT

PURPOSE: Large language models (LLMs) have proven performance for certain diagnostic tasks; however, limited studies have evaluated their consistency in recommending appropriate medication regimens for a given diagnosis. Medication management is a complex task that requires synthesis of drug formulation and complete order instructions for safe use. Here, the performance of GPT-4o, an LLM available with OpenAI’s ChatGPT, was tested on 3 medication management tasks.

METHODS: GPT-4o performance was tested on 3 medication tasks: identifying available formulations for a given generic drug name, identifying drug-drug interactions (DDIs) for a given medication regimen, and preparing a medication order for a given generic drug name. For each experiment, the model’s raw text response was captured exactly as returned and evaluated using clinician evaluation in addition to standard LLM metrics, including Term Frequency-Inverse Document Frequency (TF-IDF) vectors, normalized Levenshtein similarity, and Recall-Oriented Understudy for Gisting Evaluation (ROUGE-1/ROUGE-L) F1 score between each response and its reference string.

RESULTS: For the first task of drug-formulation matching, GPT-4o had 49% accuracy for generic medications being matched to all available formulations, with an average of 1.23 omissions per medication and 1.14 hallucinations per medication. For the second task of drug-drug interaction identification, the accuracy was 54.7% for identifying the DDI pair. For the third task, GPT-4o generated order sentences containing no medication or abbreviation errors in 65.8% of the cases.

CONCLUSION: Model performance for basic medication tasks was consistently poor. This evaluation highlights the need for domain-specific training through clinician-annotated datasets and a comprehensive evaluation framework for benchmarking performance.

PMID:42717660 | DOI:10.1093/ajhp/zxag282

EXCLI J. 2026 Jul 27;25:1215-1218. doi: 10.17179/excli2026-9686. eCollection 2026.

ABSTRACT

[This corrects the article on p. 1107 in vol. 25, PMID: 42517087.].

PMID:42712687 | PMC:PMC13550661 | DOI:10.17179/excli2026-9686

Front Plant Sci. 2026 Aug 26;17:1893113. doi: 10.3389/fpls.2026.1893113. eCollection 2026.

ABSTRACT

Winter wheat plays a viable role in agriculture, not only as a primary grain crop but also as a valuable forage source that bridges fall-spring forage gaps in many regions, including the southeastern (SE) U.S. Despite its nutritive potential, genetic basis of forage-quality traits remains insufficiently understood, limiting breeding efforts for dual-purpose cultivars. This study aimed to dissect the genetic architecture of forage quality in 182 soft red winter wheat (SRWW) genotypes adapted to the SE U.S. using genome-wide association study (GWAS). Field experiments were carried out in randomized complete block design across three Georgia locations over two growing seasons (2023-2025), with forage sampled at the end of tillering and evaluated using near-infrared reflectance spectroscopy. Significant phenotypic variation was observed for dry matter (DM), crude protein (CP), acid and neutral detergent fiber (ADF, NDF), acid detergent lignin (ADL), total digestible nutrients (TDN), sugars (SUG), and relative forage quality (RFQ). Heritability estimates ranged from low-to-moderate in combined environments and from low-to-high within individual locations. Correlation analysis revealed strong positive associations among fiber-related traits and negative associations with TDN, RFQ, and SUG, while CP declined with increasing fiber. Genome-wide association analysis identified 282 significant marker-trait associations (P < 1×10^-4) across 19 chromosomes, which were consolidated into 121 QTLs, including 27 major-effect QTLs. Three QTLs QRfq.uga-3B.1, QRfq.uga-3B.2 (RFQ) and QDm/Sug.uga-7A (DM, SUG) were stable across locations while QAdf/Adl.uga-2A (ADF, ADL) and QDm/Sug.uga-7A (DM, SUG) indicated multi-trait control. Notably, 25 of the 27 major QTLs were putatively novel, highlighting substantial untapped allelic diversity for forage-quality improvement in SE SRWW. Favorable allele accumulation resulted in an overall improvement in forage quality, increasing desirable nutritive traits (DM, RFQ, SUG, CP) while reducing undesirable traits (ADF, ADL). Candidate gene analysis linked six major QTLs with genes implicated in abiotic stress response, plant development, and metabolic regulation, supporting their functional relevance in forage-quality determination. Incorporating these loci into breeding programs provides a robust genetic framework for marker-assisted selection, enabling the development of dual-purpose wheat cultivars with enhanced forage quality, thereby strengthening wheat’s utility as a reliable forage resource during periods of seasonal feed scarcity in SE production systems.

PMID:42718753 | PMC:PMC13553917 | DOI:10.3389/fpls.2026.1893113

New Phytol. 2026 Sep 9. doi: 10.1111/nph.71535. Online ahead of print.

ABSTRACT

Repairing damaged tissues is essential for the survival of all organisms. In plants, tissue injury rapidly triggers defense and repair programs. However, the molecular mechanisms linking early injury cues to the later stage of wound repair remain unclear. Here, we show that wounding of Arabidopsis leaves induces localized low temperature at the injury site, likely caused by evaporative cooling, which is accompanied by an activation of cold-responsive genes. Using thermal imaging combined with computer vision and deep learning, we developed a workflow to monitor the dynamics of wound healing in a quantitative, non-invasive, and real-time manner. Mechanistically, we show that C-repeat Binding Factor (CBF) transcription factors are required for the activation of the injury-associated cold response and downstream salicylic acid (SA) signaling. Our findings suggest that the CBF-SA pathway acts coordinately to promote lignin and callose deposition, thereby facilitating wound repair. Together, these findings reveal a link between a wound-induced biophysical cue and the tissue repair program.

PMID:42717275 | DOI:10.1111/nph.71535

Phytopathology. 2026 Sep 9. doi: 10.1094/PHYTO-05-26-0179-R. Online ahead of print.

ABSTRACT

The soybean cyst nematode (SCN; Heterodera glycines) poses a major challenge to soybean production, intensified by the declining effectiveness of natural resistance against this pathogen. Although the use of resistant soybean varieties can be effective, their widespread and repeated use ultimately results in the emergence of virulent nematode populations that can successfully attack these resistant hosts. To assess for potential trade-offs between virulence and fitness, we investigated the hatch response, penetration rate, and reproductive potential of SCN adapted to overcome the broad-spectrum resistance source PI 437654. The hatching process is a critical phase in the life cycle of the SCN, influencing its ability to infect hosts and complete its life cycle. Our results indicated that SCN populations exhibit preferential and heightened hatch responses to their adapted host compared to alternative hosts, regardless of their virulence profile. Additionally, we found that SCN populations adapted to overcome broad-spectrum resistance showed reduced reproductive success on susceptible hosts compared to unadapted populations. This reduction in reproductive success was not attributed to differences in hatch response or penetration rates. The results from our study highlight the potential trade-offs associated with SCN virulence adaptation and emphasize the importance of considering these evolutionary dynamics in developing sustainable management strategies.

PMID:42714155 | DOI:10.1094/PHYTO-05-26-0179-R

Ecology. 2026 Sep;107(9):e70478. doi: 10.1002/ecy.70478.

ABSTRACT

The frequency and severity of droughts are increasing in grasslands globally. Forecasting changes in community composition is challenging due to the direct and indirect effects of drought as well as varied species ecologies. Rare species are predicted to be more susceptible to perturbations, but extreme drought may also disproportionately reduce common species that are well adapted to modal environmental conditions. Resolving these contrasting effects is key to predicting species reordering in response to extreme drought. We explored how species drought responses varied by their spatial and temporal rarity, defined by their local abundances relative to co-occurring species, at six sites across a gradient of mean annual precipitation (MAP) in the Central USA. We found that in general, both spatially sparse and temporally intermittent species increased during and after drought, while common and persistent species declined. Responses were similar at four of the six sites across a gradient of MAP, but did not align with the precipitation gradient as expected. Instead, patterns depended on the characteristics of common species, where rarity was not related to drought responses at sites with many common, early-season, C₃ annual grasses that likely avoided the effects of the summer drought. Finally, the increase in rare species persisted for 4 years after the experimental drought was lifted, suggesting that this extreme drought resulted in lasting shifts to community composition. These findings highlight that rare species can be favored under non-modal environmental conditions like extreme drought and underscore the importance of assessing species recovery to periodic environmental disturbances.

PMID:42710502 | DOI:10.1002/ecy.70478

Tree Physiol. 2026 Sep 8:tpag128. doi: 10.1093/treephys/tpag128. Online ahead of print.

ABSTRACT

In hybrid plants, phenotypic outcomes are governed by interactions between the two parental genomes. However, the mechanisms underlying the interplay of divergent regulatory networks from these genomes remain poorly understood. In this study, we compared gene-level and allele-specific expression patterns, as well as differentially enriched pathways between F₁ and complex backcross (CBC) lines derived from a natural interspecific hybrid population of Populus fremontii (Pf) and P. angustifolia (Pa). Metabolic differences between Pf and Pa which exhibit low and high levels respectively of phenylpropanoid-derived condensed tannins were leveraged. Using individualized transcriptome references, differential expression and clustering analyses revealed CBC-biased and F₁-biased expression for genes involved in phenylpropanoid metabolism and photosynthesis, respectively. Biased expression of these genes at the allele level was also observed in F1. At the whole-transcriptome level, Pa-biased genes predominated in F₁ hybrids, and Pa alleles displayed more conserved expression patterns than Pf alleles across examined samples. Further analyses indicated that allelic expression bias was significantly associated with parental origin, which could be driven by sequence variations in cis-regulatory elements and differences in CpG island length. Our findings demonstrate strong parent-of-origin effects on divergent regulatory networks governing gene expression in poplar hybrids and provide clues for strategic parental selection tailored to specific metabolic pathways of interest.

PMID:42708896 | DOI:10.1093/treephys/tpag128

Theor Appl Genet. 2026 Sep 5;139(9):259. doi: 10.1007/s00122-026-05326-2.

ABSTRACT

Tomato fruit weight is primarily controlled by stable genetic effects and domestication-/improvement-associated loci, including prominent chromosome 5 signals, and integrating SNP, INDEL, and SV diversity improves candidate-locus discovery, biological interpretation, and genomic prediction across environments. Tomato fruit weight (fw) is a major breeding target shaped by domestication, crop improvement, and environmental variation. We investigated the genetic architecture and genomic predictability of fw in the Varitome population representing the tomato domestication continuum, including Solanum pimpinellifolium (SP), S. lycopersicum var. cerasiforme (SLC), and cultivated S. lycopersicum (SLL). Genome-wide SNPs, insertions and deletions (INDELs), and structural variants (SVs) were analyzed individually and in combination to assess their contributions to association mapping, candidate locus discovery, and genomic prediction. Population structure based on all three variant classes clearly separated the domestication groups. Genome-wide association analyses identified 15 SNP, 10 INDEL, and 10 SV loci significantly associated with fw across environments. Several associations co-localized with known fw genes, including fw2.2/CNR, fw3.2/KLUH, fw11.3/CSR, lc/WUSCHEL, and fas/CLAVA3 supporting the biological relevance of the detected signals. Chromosome (Chr) 5 was particularly notable, with a 390-bp deletion at chr5_45,551,023 detected across all four environments and an INDEL at chr5_55,361,233 detected across three environments, suggesting stable improvement-associated candidate loci for fw variation validation. Multiple loci exhibited clear allele-frequency shifts from SP through SLC to SLL, consistent with selection during domestication and improvement, whereas others were detected almost exclusively in cultivated germplasm, suggesting more recent breeding-associated origins. Fw displayed strong genetic control, with genotype explaining more than 92% of the phenotypic variance and only minor contributions from environment and genotype-by-environment interactions. Five genomic prediction models based on SNPs, INDELs, SVs, their combinations, and genotype-by-environment interaction kernels were evaluated using four cross-validation scenarios (CV1, CV2, CV0, and CV00). Prediction accuracy was influenced more strongly by validation scenario and prediction algorithm choice than by marker class. Partial Least Squares (PLS) and Bayesian Genomic Linear Regression (BGLR) consistently outperformed Random Forest (RF) and Deep Learning (DL) models, particularly under scenarios involving untested genotypes and environments. INDEL-based models frequently achieved the highest accuracies under the most stringent prediction scenarios, although differences among marker classes were generally modest. Integrating multiple classes of genomic variation improved biological interpretation, enabled the identification of domestication- and improvement-associated loci, and provided accurate prediction across environments and genetic backgrounds. Together, these results demonstrate that tomato fw is governed primarily by stable additive genetic effects and highlight the value of multi-variant genomic approaches for both genetic dissection and genomic-assisted breeding.

PMID:42700247 | PMC:PMC13546169 | DOI:10.1007/s00122-026-05326-2

J Invertebr Pathol. 2026 Sep 6:108742. doi: 10.1016/j.jip.2026.108742. Online ahead of print.

ABSTRACT

Oil-in-water Pickering emulsions stabilized by titania (TiO2) nanoparticles have shown potential to increase ultraviolet (UV) protection in the entomopathogenic fungus Metarhizium brunneum. In this study, we expand the findings by comparing several nanoparticle-based formulations for UV tolerance, and investigating the impact of protective formulations in a biocontrol setting. The fungal viability in colony forming units (CFUs) and virulence to the 2nd instar tobacco thrips, Frankliniella fusca, were assessed. Under UV-A (380-nm), a titania-based formulation was effective in protecting the fungal viability at 6-h exposure but not at 12-h; silica (SiO2) did not provide any UV protection. UV-A radiation for up to 12 h did not reduce the fungal virulence in any treatment; then a stronger UV treatment was warranted. Under UV-C (254-nm), silica was ineffective in improving the viability or virulence of M. brunneum. The titania-based emulsion showed partial protection for up to 10-min exposure, with 51-56% reduction in CFU counts compared to nearly 100% reduction in the control, while it only exhibited marginal effect in assessing fungal virulence. Two titania-based formulations (titania-MO-1 and titania-PO-2) were further explored for enhanced UV tolerance. Titania- MO-1 had fewer CFUs at 10-min UV compared to control; titania- PO-2 showed no reduction in fungal viability. The two titania formulations were equally effective in maintaining fungal virulence to F. fusca regardless of UV exposure. Overall, titania nanoparticles (particularly titania- PO-2) protected M. brunneum conidia against UV radiation, suggesting the potential use of titania nanoparticles in formulating M. brunneum for improved UV tolerance and biocontrol potential.

PMID:42702346 | DOI:10.1016/j.jip.2026.108742

Theor Appl Genet. 2026 Sep 5;139(9):260. doi: 10.1007/s00122-026-05348-w.

ABSTRACT

We mapped key regions of the maize genome that influence the formation of aboveground (aerial) roots, which in some varieties have been associated with symbiosis with nitrogen-fixing bacteria. Modern agriculture relies heavily on chemically synthesized nitrogen fertilizers, which ensure high yields but also carry high economic and environmental costs. Biological nitrogen fixation (BNF) supplies high amounts of nitrogen to legumes, and several avenues of research are underway to extend it to cereal crops. In maize, aerial roots formed in Sierra Mixe landraces have been associated with BNF. However, much of the genetics underlying aerial root morphology remains unknown. Here, we evaluate aerial root morphology traits associated with BNF in three segregating populations derived from crosses between two Midwest-adapted inbred lines and three landraces. Inclusive composite interval mapping (iCIM) with flowering time as a covariate identified 37 quantitative trait loci (QTL) for three aerial root traits (nodes with roots, root size, and roots per node) which exhibit moderately high heritability (H² = 0.65 to 0.83). The combined proportion of phenotypic variance explained by the detected QTL ranged from 23 to 51%, depending on the trait and population, and potential candidate genes were identified through literature searches, macrosynteny, and gene expression analyses. Introgressing the most relevant aerial root-associated QTL into elite genotypes may provide a path toward achieving meaningful levels of BNF-associated traits in maize, but further work is needed to assess this approach’s viability under field conditions.

PMID:42700241 | DOI:10.1007/s00122-026-05348-w