Generational Specific Recruitment of Arsenic- and Antimony-Reducing Microorganisms in Plant Root-Associated Niches for Adapting to Metalloid-Metal Pollution.

The recruitment of microorganisms by plants can enhance their adaptability to environmental stressors, but how root-associated niches recruit specific microorganisms for adapting to metalloid-metal contamination is not well-understood. This study investigated the generational recruitment of microorganisms in different root niches of Vetiveria zizanioides (V. zizanioides) under arsenic (As) and antimony (Sb) stress. The V. zizanioides was cultivated in As- and Sb-cocontaminated mine soils (MS) and artificial pollution soils (PS) over two generations in controlled conditions. The root-associated microbial communities were analyzed through 16S rRNA, arsC, and aioA gene amplicon and metagenomics sequencing. V. zizanioides accumulated higher As(III) and Sb(III) in its endosphere in MS in the second generation, while its physiological indices in MS were better than those observed in PS. SourceTracker analysis revealed that V. zizanioides in MS recruited As(V)- and Sb(V)-reducing microorganisms (e.g., Sphingomonales and Rhodospirillaceae) into the rhizoplane and endosphere. Metagenomics analysis further confirmed that these recruited microorganisms carrying genes encoding arsenate reductases with diverse carbohydrate degradation abilities were enriched in the rhizoplane and endosphere, suggesting their potential to reduce As(V) and Sb(V) and to decompose root exudates (e.g., xylan and starch). These findings reveal that V. zizanioides selectively recruits As- and Sb-reducing microorganisms to mitigate As-Sb cocontamination during the generational growth, providing insights into novel strategies for enhancing phytoremediation of metalloid-metal contaminants.

SELECTOR: Heterogeneous graph network with convolutional masked autoencoder for multimodal robust prediction of cancer survival.

Accurately predicting the survival rate of cancer patients is crucial for aiding clinicians in planning appropriate treatment, reducing cancer-related medical expenses, and significantly enhancing patients' quality of life. Multimodal prediction of cancer patient survival offers a more comprehensive and precise approach. However, existing methods still grapple with challenges related to missing multimodal data and information interaction within modalities. This paper introduces SELECTOR, a heterogeneous graph-aware network based on convolutional mask encoders for robust multimodal prediction of cancer patient survival. SELECTOR comprises feature edge reconstruction, convolutional mask encoder, feature cross-fusion, and multimodal survival prediction modules. Initially, we construct a multimodal heterogeneous graph and employ the meta-path method for feature edge reconstruction, ensuring comprehensive incorporation of feature information from graph edges and effective embedding of nodes. To mitigate the impact of missing features within the modality on prediction accuracy, we devised a convolutional masked autoencoder (CMAE) to process the heterogeneous graph post-feature reconstruction. Subsequently, the feature cross-fusion module facilitates communication between modalities, ensuring that output features encompass all features of the modality and relevant information from other modalities. Extensive experiments and analysis on six cancer datasets from TCGA demonstrate that our method significantly outperforms state-of-the-art methods in both modality-missing and intra-modality information-confirmed cases. Our codes are made available at https://github.com/panliangrui/Selector.