AMAR-seq: Automated Multimodal Sequencing of DNA Methylation, Chromatin Accessibility, and RNA Expression with Single-Cell Resolution.

Parallel single-cell multimodal sequencing is the most intuitive and precise tool for cellular status research. In this study, we propose AMAR-seq to automate methylation, chromatin accessibility, and RNA expression coanalysis with single-cell precision. We validated the accuracy and robustness of AMAR-seq in comparison with standard single-omics methods. The high gene detection rate and genome coverage of AMAR-seq enabled us to establish a genome-wide gene expression regulatory atlas and triple-omics landscape with single base resolution and implement single-cell copy number variation analysis. Applying AMAR-seq to investigate the process of mouse embryonic stem cell differentiation, we revealed the dynamic coupling of the epigenome and transcriptome, which may contribute to unraveling the molecular mechanisms of early embryonic development. Collectively, we propose AMAR-seq for the in-depth and accurate establishment of single-cell multiomics regulatory patterns in a cost-effective, efficient, and automated manner, paving the way for insightful dissection of complex life processes.

Domain generalization enables general cancer cell annotation in single-cell and spatial transcriptomics.

Single-cell and spatial transcriptome sequencing, two recently optimized transcriptome sequencing methods, are increasingly used to study cancer and related diseases. Cell annotation, particularly for malignant cell annotation, is essential and crucial for in-depth analyses in these studies. However, current algorithms lack accuracy and generalization, making it difficult to consistently and rapidly infer malignant cells from pan-cancer data. To address this issue, we present Cancer-Finder, a domain generalization-based deep-learning algorithm that can rapidly identify malignant cells in single-cell data with an average accuracy of 95.16%. More importantly, by replacing the single-cell training data with spatial transcriptomic datasets, Cancer-Finder can accurately identify malignant spots on spatial slides. Applying Cancer-Finder to 5 clear cell renal cell carcinoma spatial transcriptomic samples, Cancer-Finder demonstrates a good ability to identify malignant spots and identifies a gene signature consisting of 10 genes that are significantly co-localized and enriched at the tumor-normal interface and have a strong correlation with the prognosis of clear cell renal cell carcinoma patients. In conclusion, Cancer-Finder is an efficient and extensible tool for malignant cell annotation.

DIRECT: Digital Microfluidics for Isolation-Free Shared Library Construction of Single-Cell DNA Methylome and Transcriptome.

Simultaneous profiling of DNA methylation and gene expression within single cells is a powerful technology to dissect complex gene regulatory network of cells. However, existing methods are based on picking a single-cell in a tube and split single-cell lysate into two parts for transcriptome and methylome library construction, respectively, which is costly and cumbersome. Here, DIRECT is proposed, a digital microfluidics-based method for high-efficiency single-cell isolation and simultaneous analysis of the methylome and transcriptome in a single library construction. The accuracy of DIRECT is demonstrated in comparison with bulk and single-omics data, and the high CpG site coverage of DIRECT allows for precise analysis of copy number variation information, enabling expansion of single cell analysis from two- to three-omics. By applying DIRECT to monitor the dynamics of mouse embryonic stem cell differentiation, the relationship between DNA methylation and changes in gene expression during differentiation is revealed. DIRECT enables accurate, robust, and reproducible single-cell DNA methylation and gene expression co-analysis in a more cost-effective, simpler library preparation and automated manner, broadening the application scenarios of single-cell multi-omics analysis and revealing a more comprehensive and fine-grained map of cellular regulatory landscapes.

Digital-scRRBS: A Cost-Effective, Highly Sensitive, and Automated Single-Cell Methylome Analysis Platform via Digital Microfluidics.

Single-cell DNA methylation sequencing is highly effective for identifying cell subpopulations and constructing epigenetic regulatory networks. Existing methylome analyses require extensive starting materials and are costly, complex, and susceptible to contamination, thereby impeding the development of single-cell methylome technology. In this work, we report digital microfluidics-based single-cell reduced representation bisulfite sequencing (digital-scRRBS), the first microfluidics-based single-cell methylome library construction platform, which is an automatic, effective, reproducible, and reagent-efficient technique to dissect the single-cell methylome. Using our digital microfluidic chip, we isolated single cells in 15 s and successfully constructed single-cell methylation sequencing libraries with a unique genome mapping rate of up to 53.6%, covering up to 2.26 million CpG sites. Digital-scRRBS demonstrates a high capacity for distinguishing cell identity and tracking DNA methylation during drug administration. Digital-scRRBS expands the applicability of single-cell methylation methods as a versatile tool for epigenetic analysis of rare cells and populations with high levels of heterogeneity.

Deep transfer learning enables lesion tracing of circulating tumor cells.

Liquid biopsy offers great promise for noninvasive cancer diagnostics, while the lack of adequate target characterization and analysis hinders its wide application. Single-cell RNA sequencing (scRNA-seq) is a powerful technology for cell characterization. Integrating scRNA-seq into a CTC-focused liquid biopsy study can perhaps classify CTCs by their original lesions. However, the lack of CTC scRNA-seq data accumulation and prior knowledge hinders further development. Therefore, we design CTC-Tracer, a transfer learning-based algorithm, to correct the distributional shift between primary cancer cells and CTCs to transfer lesion labels from the primary cancer cell atlas to CTCs. The robustness and accuracy of CTC-Tracer are validated by 8 individual standard datasets. We apply CTC-Tracer on a complex dataset consisting of RNA-seq profiles of single CTCs, CTC clusters from a BRCA patient, and two xenografts, and demonstrate that CTC-Tracer has potential in knowledge transfer between different types of RNA-seq data of lesions and CTCs.