12 November 2026 in bioRxiv
JDJanithya Dishani Kodituwakku Dharmawardhane
Janithya Dishani Kodituwakku Dharmawardhane
+6
Hotter droughts driven by global warming are increasing tree mortality worldwide. Here, we propose a theoretical framework that explains how dissolved gas affects the stability of bubbles in xylem sap, and shows that in the presence of gas, sudden increases in temperature can accelerate embolism formation. We then provide experimental evidence that heat exposure can change xylem vulnerability to drought. Using data from seven Australian and European tree species, we assessed how elevated temperature (+15 and +20°C above ambient) affects both xylem embolism resistance, measured with the Optical Vulnerability Technique, and VPD-corrected drying times. We found evidence for reduced embolism resistance and accelerated dehydration in heat-exposed plants. On average, trees reached the water potential at 50% loss of xylem function at 0.73 MPa less negative water potentials and 7.8 kPa h earlier, though responses were highly species-specific, and credibly different from zero only in the Australian species. Our results are consistent with the proposed framework and suggest that embolism resistance may not be a static trait, which has important repercussions for studies of plant vulnerability to drought. Further studies are necessary to assess the potential effects of heat-induced reductions in embolism resistance on tree mortality during heatwaves.
24 October 2026 in bioRxiv
Motivation: Many spatial domain identification methods describe the neighbourhood of a cell by mean gene expression. However, the mean cannot separate neighbourhoods that differ only in gene–gene covariance and tissue is organised at several spatial scales.ht Results: We propose CoSMIM, which represents the niche of each cell by a weighted fusion of gene–gene covariance matrices of its spatial neighbourhood at multiple scales. Principal component analysis and Leiden clustering of the matrix square root of the fused matrix identify spatial domains without model training. On six benchmark datasets from five platforms, CoSMIM attains the best overall rank over eight evaluation metrics among 16 methods. It also ranks first on a Xenium mouse brain dataset, and on all seven datasets its median NMI exceeds that of single-scale COVET. The domain covariance matrices also show gene pairs whose correlation differs between domains. Availability and implementation: The source code, documentation and test data of CoSMIM are available at GitHub ( https://github.com/barondu/cosmim ). Contact: du_bairui_24@stu-cbms.k.u-tokyo.ac.jp Supplementary information: Supplementary data are available at Bioinformatics online.