Dictionary learning for integrative, multimodal and scalable single-cell analysis

Nature Biotechnology, Год журнала: 2023, Номер 42(2), С. 293 - 304

Опубликована: Май 25, 2023

Язык: Английский

---

Benchmarking atlas-level data integration in single-cell genomics

Nature Methods, Год журнала: 2021, Номер 19(1), С. 41 - 50

Опубликована: Дек. 23, 2021

Abstract Single-cell atlases often include samples that span locations, laboratories and conditions, leading to complex, nested batch effects in data. Thus, joint analysis of atlas datasets requires reliable data integration. To guide integration method choice, we benchmarked 68 preprocessing combinations on 85 batches gene expression, chromatin accessibility simulation from 23 publications, altogether representing >1.2 million cells distributed 13 atlas-level tasks. We evaluated methods according scalability, usability their ability remove while retaining biological variation using 14 evaluation metrics. show highly variable selection improves the performance methods, whereas scaling pushes prioritize removal over conservation variation. Overall, scANVI, Scanorama, scVI scGen perform well, particularly complex tasks, single-cell ATAC-sequencing is strongly affected by choice feature space. Our freely available Python module benchmarking pipeline can identify optimal for new data, benchmark improve development.

Язык: Английский

---

The Tabula Sapiens: A multiple-organ, single-cell transcriptomic atlas of humans

Science, Год журнала: 2022, Номер 376(6594)

Опубликована: Май 12, 2022

Molecular characterization of cell types using single-cell transcriptome sequencing is revolutionizing biology and enabling new insights into the physiology human organs. We created a reference atlas comprising nearly 500,000 cells from 24 different tissues organs, many same donor. This enabled molecular more than 400 types, their distribution across tissues, tissue-specific variation in gene expression. Using multiple single donor identification clonal T between mutation rate B cells, analysis cycle state proliferative potential shared tissues. Cell type–specific RNA splicing was discovered analyzed within an individual.

Язык: Английский

---

Best practices for single-cell analysis across modalities

Nature Reviews Genetics, Год журнала: 2023, Номер 24(8), С. 550 - 572

Опубликована: Март 31, 2023

Язык: Английский

---

A Python library for probabilistic analysis of single-cell omics data

Nature Biotechnology, Год журнала: 2022, Номер 40(2), С. 163 - 166

Опубликована: Фев. 1, 2022

Язык: Английский

---

Mapping single-cell data to reference atlases by transfer learning

Nature Biotechnology, Год журнала: 2021, Номер 40(1), С. 121 - 130

Опубликована: Авг. 30, 2021

Abstract Large single-cell atlases are now routinely generated to serve as references for analysis of smaller-scale studies. Yet learning from reference data is complicated by batch effects between datasets, limited availability computational resources and sharing restrictions on raw data. Here we introduce a deep strategy mapping query datasets top called architectural surgery (scArches). scArches uses transfer parameter optimization enable efficient, decentralized, iterative building contextualization new with existing without Using examples mouse brain, pancreas, immune whole-organism atlases, show that preserves biological state information while removing effects, despite using four orders magnitude fewer parameters than de novo integration. generalizes multimodal mapping, allowing imputation missing modalities. Finally, retains coronavirus disease 2019 (COVID-19) variation when healthy reference, enabling the discovery disease-specific cell states. will facilitate collaborative projects construction, updating, efficient use atlases.

Язык: Английский

---

Multi-omics single-cell data integration and regulatory inference with graph-linked embedding

Nature Biotechnology, Год журнала: 2022, Номер 40(10), С. 1458 - 1466

Опубликована: Май 2, 2022

Despite the emergence of experimental methods for simultaneous measurement multiple omics modalities in single cells, most single-cell datasets include only one modality. A major obstacle integrating data from is that different layers typically have distinct feature spaces. Here, we propose a computational framework called GLUE (graph-linked unified embedding), which bridges gap by modeling regulatory interactions across explicitly. Systematic benchmarking demonstrated more accurate, robust and scalable than state-of-the-art tools heterogeneous multi-omics data. We applied to various challenging tasks, including triple-omics integration, integrative inference human cell atlas construction over millions where was able correct previous annotations. features modular design can be flexibly extended enhanced new analysis tasks. The full package available online at https://github.com/gao-lab/GLUE .

Язык: Английский

---

High-resolution single-cell atlas reveals diversity and plasticity of tissue-resident neutrophils in non-small cell lung cancer

Cancer Cell, Год журнала: 2022, Номер 40(12), С. 1503 - 1520.e8

Опубликована: Ноя. 10, 2022

Non-small cell lung cancer (NSCLC) is characterized by molecular heterogeneity with diverse immune infiltration patterns, which has been linked to therapy sensitivity and resistance. However, full understanding of how phenotypes vary across different patient subgroups lacking. Here, we dissect the NSCLC tumor microenvironment at high resolution integrating 1,283,972 single cells from 556 samples 318 patients 29 datasets, including our dataset capturing low mRNA content. We stratify into immune-deserted, B cell, T myeloid subtypes. Using bulk genomic clinical information, identify cellular components associated histology genotypes. then focus on analysis tissue-resident neutrophils (TRNs) uncover distinct subpopulations that acquire new functional properties in tissue microenvironment, providing evidence for plasticity TRNs. Finally, show a TRN-derived gene signature anti-programmed death ligand 1 (PD-L1) treatment failure.

Язык: Английский

---

Functional interpretation of single cell similarity maps

Nature Communications, Год журнала: 2019, Номер 10(1)

Опубликована: Сен. 26, 2019

We present Vision, a tool for annotating the sources of variation in single cell RNA-seq data an automated and scalable manner. Vision operates directly on manifold cell-cell similarity employs flexible annotation approach that can operate either with or without preconceived stratification cells into groups along continuum. demonstrate utility several case studies show it derive important cellular link them to experimental meta-data even relatively homogeneous sets cells. produces interactive, low latency feature rich web-based report be easily shared among researchers, thus facilitating dissemination collaboration.

Язык: Английский

---

Lineage tracing reveals the phylodynamics, plasticity, and paths of tumor evolution

Cell, Год журнала: 2022, Номер 185(11), С. 1905 - 1923.e25

Опубликована: Май 1, 2022

Tumor evolution is driven by the progressive acquisition of genetic and epigenetic alterations that enable uncontrolled growth expansion to neighboring distal tissues. The study phylogenetic relationships between cancer cells provides key insights into these processes. Here, we introduced an evolving lineage-tracing system with a single-cell RNA-seq readout mouse model Kras;Trp53(KP)-driven lung adenocarcinoma tracked tumor from single-transformed metastatic tumors at unprecedented resolution. We found loss initial, stable alveolar-type2-like state was accompanied transient increase in plasticity. This followed adoption distinct transcriptional programs rapid and, ultimately, clonal sweep subclones capable metastasizing. Finally, develop through stereotypical evolutionary trajectories, perturbing additional suppressors accelerates progression creating novel trajectories. Our elucidates hierarchical nature more broadly, enables in-depth studies progression.

Язык: Английский

---