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

---

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

---

The specious art of single-cell genomics

PLoS Computational Biology, Год журнала: 2023, Номер 19(8), С. e1011288 - e1011288

Опубликована: Авг. 17, 2023

Dimensionality reduction is standard practice for filtering noise and identifying relevant features in large-scale data analyses. In biology, single-cell genomics studies typically begin with to 2 or 3 dimensions produce "all-in-one" visuals of the that are amenable human eye, these subsequently used qualitative quantitative exploratory analysis. However, there little theoretical support this practice, we show extreme dimension reduction, from hundreds thousands 2, inevitably induces significant distortion high-dimensional datasets. We therefore examine practical implications low-dimensional embedding find extensive distortions inconsistent practices make such embeddings counter-productive exploratory, biological lieu this, discuss alternative approaches conducting targeted feature exploration enable hypothesis-driven discovery.

Язык: Английский

---

Characterizing cis-regulatory elements using single-cell epigenomics

Nature Reviews Genetics, Год журнала: 2022, Номер 24(1), С. 21 - 43

Опубликована: Июль 15, 2022

Язык: Английский

---

Platelet-instructed SPP1+ macrophages drive myofibroblast activation in fibrosis in a CXCL4-dependent manner

Cell Reports, Год журнала: 2023, Номер 42(2), С. 112131 - 112131

Опубликована: Фев. 1, 2023

Fibrosis represents the common end stage of chronic organ injury independent initial insult, destroying tissue architecture and driving failure. Here we discover a population profibrotic macrophages marked by expression Spp1, Fn1, Arg1 (termed Spp1 macrophages), which expands after injury. Using an unbiased approach, identify chemokine (C-X-C motif) ligand 4 (CXCL4) to be among top upregulated genes during macrophage differentiation. In vitro in vivo studies show that loss Cxcl4 abrogates differentiation ameliorates fibrosis both heart kidney Moreover, find platelets, most abundant source CXCL4 vivo, drive Single nuclear RNA sequencing with ligand-receptor interaction analysis reveals orchestrate fibroblast activation via Sema3 crosstalk. Finally, confirm expand human disease

Язык: Английский

---

Cross-tissue, single-cell stromal atlas identifies shared pathological fibroblast phenotypes in four chronic inflammatory diseases

Med, Год журнала: 2022, Номер 3(7), С. 481 - 518.e14

Опубликована: Май 31, 2022

Pro-inflammatory fibroblasts are critical for pathogenesis in rheumatoid arthritis, inflammatory bowel disease, interstitial lung and Sjögren's syndrome represent a novel therapeutic target chronic disease. However, the heterogeneity of fibroblast phenotypes, exacerbated by lack common cross-tissue taxonomy, has limited our understanding which pathways shared multiple diseases.

Язык: Английский

---

Human prefrontal cortex gene regulatory dynamics from gestation to adulthood at single-cell resolution

Cell, Год журнала: 2022, Номер 185(23), С. 4428 - 4447.e28

Опубликована: Окт. 31, 2022

Human brain development is underpinned by cellular and molecular reconfigurations continuing into the third decade of life. To reveal cell dynamics orchestrating neural maturation, we profiled human prefrontal cortex gene expression chromatin accessibility at single-cell resolution from gestation to adulthood. Integrative analyses define dynamic trajectories each type, revealing major reconfiguration prenatal-to-postnatal transition in all types followed continuous adulthood identifying regulatory networks guiding developmental programs, states, functions. We uncover links between milestones, characterize diverse timing when cells acquire adult-like identify convergence distinct origins. further their regulators implicated neurological disorders. Finally, using this reference, benchmark identities maturation states organoid models. Together, captures landscape cortical development.

Язык: Английский

---

The Specious Art of Single-Cell Genomics

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2021, Номер unknown

Опубликована: Авг. 26, 2021

Abstract Dimensionality reduction is standard practice for filtering noise and identifying relevant features in large-scale data analyses. In biology, single-cell genomics studies typically begin with to two or three dimensions produce ‘all-in-one’ visuals of the that are amenable human eye, these subsequently used qualitative quantitative exploratory analysis. However, there little theoretical support this practice, we show extreme dimension reduction, from hundreds thousands two, inevitably induces significant distortion high-dimensional datasets. We therefore examine practical implications low-dimensional embedding data, find extensive distortions inconsistent practices make such embeddings counter-productive exploratory, biological lieu this, discuss alternative approaches conducting targeted feature exploration, enable hypothesis-driven discovery.

Язык: Английский

---

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

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2022, Номер unknown

Опубликована: Фев. 26, 2022

Abstract Mapping single-cell sequencing profiles to comprehensive reference datasets represents a powerful alternative unsupervised analysis. Reference datasets, however, are predominantly constructed from RNA-seq data, and cannot be used annotate that do not measure gene expression. Here we introduce ‘bridge integration’, method harmonize singlecell across modalities by leveraging multi-omic dataset as molecular bridge. Each cell in the comprises an element ‘dictionary’, which can reconstruct unimodal transform them into shared space. We demonstrate our procedure accurately transcriptomic data with independent single measurements of chromatin accessibility, histone modifications, DNA methylation, protein levels. Moreover, how dictionary learning combined sketching techniques substantially improve computational scalability, 8.6 million human immune mass cytometry experiments. Our approach aims broaden utility facilitate comparisons diverse modalities. Availability Installation instructions, documentations, vignettes available at http://www.satijalab.org/seurat

Язык: Английский

---

Human Bone Marrow Organoids for Disease Modeling, Discovery, and Validation of Therapeutic Targets in Hematologic Malignancies

Cancer Discovery, Год журнала: 2022, Номер 13(2), С. 364 - 385

Опубликована: Ноя. 9, 2022

Abstract A lack of models that recapitulate the complexity human bone marrow has hampered mechanistic studies normal and malignant hematopoiesis validation novel therapies. Here, we describe a step-wise, directed-differentiation protocol in which organoids are generated from induced pluripotent stem cells committed to mesenchymal, endothelial, hematopoietic lineages. These 3D structures capture key features marrow—stroma, lumen-forming sinusoids, myeloid including proplatelet-forming megakaryocytes. The supported engraftment survival patients with blood malignancies, cancer types notoriously difficult maintain ex vivo. Fibrosis organoid occurred following TGFβ stimulation myelofibrosis but not healthy donor–derived cells, validating this platform as powerful tool for their interactions within marrow–like milieu. This enabling technology is likely accelerate discovery prioritization targets disorders cancers. Significance: We present supports growth primary lymphoid model allows cancers context microenvironment provides much-needed vivo new therapeutics. See related commentary by Derecka Crispino, p. 263. article highlighted In Issue feature, 247

Язык: Английский

---