Deciphering universal gene regulatory mechanisms in diverse organisms holds great potential for advancing our knowledge of fundamental life processes and facilitating clinical applications. However, the traditional research paradigm primarily focuses on individual model organisms and does not integrate various cell types across species. Recent breakthroughs in single-cell sequencing and deep learning techniques present an unprecedented opportunity to address this challenge. In this study, we built an extensive dataset of over 120 million human and mouse single-cell transcriptomes. After data preprocessing, we obtained 101,768,420 single-cell transcriptomes and developed a knowledge-informed cross-species foundation model, named GeneCompass. During pre-training, GeneCompass effectively integrated four types of prior biological knowledge to enhance our understanding of gene regulatory mechanisms in a self-supervised manner. By fine-tuning for multiple downstream tasks, GeneCompass outperformed state-of-the-art models in diverse applications for a single species and unlocked new realms of cross-species biological investigations. We also employed GeneCompass to search for key factors associated with cell fate transition and showed that the predicted candidate genes could successfully induce the differentiation of human embryonic stem cells into the gonadal fate. Overall, GeneCompass demonstrates the advantages of using artificial intelligence technology to decipher universal gene regulatory mechanisms and shows tremendous potential for accelerating the discovery of critical cell fate regulators and candidate drug targets.

In eukaryotic cells, lysosomes are distributed in the cytoplasm as individual membrane-bound compartments to degrade macromolecules and to control cellular metabolism. A fundamental yet unanswered question is whether and, if so, how individual lysosomes are organized spatially to coordinate and integrate their functions. To address this question, we analyzed their collective behavior in cultured cells using spatial statistical techniques. We found that in single cells, lysosomes maintain non-random, stable, yet distinct spatial distributions mediated by the cytoskeleton, the endoplasmic reticulum (ER), and lysosomal biogenesis. Throughout the intracellular space, lysosomes form dynamic clusters that significantly increase their interactions with endosomes. Cluster formation is associated with local increases in ER spatial density but does not depend on fusion with endosomes or spatial exclusion by mitochondria. Taken together, our findings reveal whole-cell scale spatial organization of lysosomes and provide insights into how organelle interactions are mediated and regulated across the entire intracellular space.

In recent years graphene has drawn considerable research interest for biomedical applications. However, applications of graphene in biological systems also raise concerns about its possible toxicity. Here, by using live cell imaging techniques, we investigate the effect of pristine graphene on the viability as well as stress of both nonneuronal and neuronal cells under physiological conditions. We find that graphene promotes cell adhesion and proliferation. Furthermore, we find that graphene has no detectable adverse effect on mitochondrial membrane potential and morphology, or autophagy levels in the cell, indicating that graphene does not induce cell stress. Our results highlight the potential of graphene to be used in biomedical applications by providing long-term and stable nonneural and neural interfaces.

Atherosclerosis (AS) is a common cardiovascular disease with complex pathogenesis, in which multiple pathways and their interweaving regulatory mechanism remain unclear. The primary transcription factor NF-κB plays a critical role in AS via modulating the expression of a series of inflammatory mediators under various stimuli such as cytokines, microbial antigens, and intracellular stresses. Endoplasmic reticulum (ER) stress, caused by the disrupted synthesis and secretion of protein, links inflammation, metabolic signals, and other cellular processes via the unfolded protein response (UPR). Both NF-κB and ER stress share the intersection regarding their molecular regulation and function and are regarded as critical individual contributors to AS. In this review, we summarize the multiple interactions between NF-κB and ER stress activation, including the UPR, NLRP3 inflammasome, and reactive oxygen species (ROS) generation, which have been ignored in the pathogenesis of AS. Given the multiple links between NF-κB and ER stress, we speculate that the integrated network contributes to the understanding of molecular mechanisms of AS. This review aims to provide an insight into these interactions and their underlying roles in the progression of AS, highlighting potential pharmacological targets against the atherosclerotic inflammatory process.

To solve three-dimensional structures of biological macromolecules in situ, large numbers of particles often need to be picked from cryo-electron tomograms. However, adoption of automated particle-picking methods remains limited because of their technical limitations. To overcome the limitations, we develop DeepETPicker, a deep learning model for fast and accurate picking of particles from cryo-electron tomograms. Training of DeepETPicker requires only weak supervision with low numbers of simplified labels, reducing the burden of manual annotation. The simplified labels combined with the customized and lightweight model architecture of DeepETPicker and accelerated pooling enable substantial performance improvement. When tested on simulated and real tomograms, DeepETPicker outperforms the competing state-of-the-art methods by achieving the highest overall accuracy and speed, which translate into higher authenticity and coordinates accuracy of picked particles and higher resolutions of final reconstruction maps. DeepETPicker is provided in open source with a user-friendly interface to support cryo-electron tomography in situ.