scholarly journals Transfer learning efficiently maps bone marrow cell types from mouse to human using single-cell RNA sequencing

2020 ◽  
Vol 3 (1) ◽  
Author(s):  
Patrick S. Stumpf ◽  
Xin Du ◽  
Haruka Imanishi ◽  
Yuya Kunisaki ◽  
Yuichiro Semba ◽  
...  
Keyword(s):  
Machine Learning ◽  
Bone Marrow ◽  
Single Cell ◽  
Rna Sequencing ◽  
Transfer Learning ◽  
Biomedical Research ◽  
Human Cell ◽  
Cell Types ◽  
Single Cell Rna Sequencing ◽  
Using Data

AbstractBiomedical research often involves conducting experiments on model organisms in the anticipation that the biology learnt will transfer to humans. Previous comparative studies of mouse and human tissues were limited by the use of bulk-cell material. Here we show that transfer learning—the branch of machine learning that concerns passing information from one domain to another—can be used to efficiently map bone marrow biology between species, using data obtained from single-cell RNA sequencing. We first trained a multiclass logistic regression model to recognize different cell types in mouse bone marrow achieving equivalent performance to more complex artificial neural networks. Furthermore, it was able to identify individual human bone marrow cells with 83% overall accuracy. However, some human cell types were not easily identified, indicating important differences in biology. When re-training the mouse classifier using data from human, less than 10 human cells of a given type were needed to accurately learn its representation. In some cases, human cell identities could be inferred directly from the mouse classifier via zero-shot learning. These results show how simple machine learning models can be used to reconstruct complex biology from limited data, with broad implications for biomedical research.

scholarly journals Mapping biology from mouse to man using transfer learning

2019 ◽  
Author(s):  
Patrick S. Stumpf ◽  
Doris Du ◽  
Haruka Imanishi ◽  
Yuya Kunisaki ◽  
Yuichiro Semba ◽  
...  
Keyword(s):  
Machine Learning ◽  
Transfer Learning ◽  
Biomedical Research ◽  
Human Cell ◽  
Cell Types ◽  
Human Cells ◽  
Model Organisms ◽  
Using Data ◽  
Artificial Neural Network Ann ◽  
Different Cell Types

Biomedical research often involves conducting experiments on model organisms in the anticipation that the biology learnt from these experiments will transfer to the human. Yet, it is commonly the case that biology does not transfer effectively, often for unknown reasons. Despite its importance to translational research this transfer process is not currently rigorously quantified. Here, we show that transfer learning – the branch of machine learning that concerns passing information from one domain to another – can be used to efficiently map biology from mouse to man, using the bone marrow (BM) as a representative example of a complex tissue. We first trained an artificial neural network (ANN) to accurately recognize various different cell types in mouse BM using data obtained from single-cell RNA-sequencing (scRNA-Seq) experiments. We found that this ANN, trained exclusively on mouse data, was able to identify individual human cells obtained from comparable scRNA-Seq experiments of human BM with 83% overall accuracy. However, while some human cell types were easily identified, others were not, indicating important differences in biology. To obtain a more accurate map of the human BM we then retrained the mouse ANN using scRNA-Seq data from a limited sample of human BM cells. Typically, less than 10 human cells of a given type were needed to accurately learn its representation in the updated model. In some cases, human cell identities could be inferred directly from the mouse ANN without retraining, via a process of biologically-guided zero-shot learning. These results show how machine learning can be used to reconstruct complex biology from limited data and have broad implications for biomedical research.

scholarly journals Polled Digital Cell Sorter (p-DCS): Automatic identification of hematological cell types from single cell RNA-sequencing clusters

2019 ◽  
Author(s):  
Sergii Domanskyi ◽  
Anthony Szedlak ◽  
Nathaniel T Hawkins ◽  
Jiayin Wang ◽  
Giovanni Paternostro ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Single Cell ◽  
Rna Sequencing ◽  
Statistical Significance ◽  
Cell Types ◽  
Automatic Identification ◽  
Molecular Signatures ◽  
Rna Seq ◽  
Cell Type ◽  
Single Cell Rna Sequencing

AbstractBackgroundSingle cell RNA sequencing (scRNA-seq) brings unprecedented opportunities for mapping the heterogeneity of complex cellular environments such as bone marrow, and provides insight into many cellular processes. Single cell RNA-seq, however, has a far larger fraction of missing data reported as zeros (dropouts) than traditional bulk RNA-seq. This makes difficult not only the clustering of cells, but also the assignment of the resulting clusters into predefined cell types based on known molecular signatures, such as the expression of characteristic cell surface markers.ResultsWe present a computational tool for processing single cell RNA-seq data that uses a voting algorithm to identify cells based on approval votes received by known molecular markers. Using a stochastic procedure that accounts for biases due to dropout errors and imbalances in the number of known molecular signatures for different cell types, the method computes the statistical significance of the final approval score and automatically assigns a cell type to clusters without an expert curator. We demonstrate the utility of the tool in the analysis of eight samples of bone marrow from the Human Cell Atlas. The tool provides a systematic identification of cell types in bone marrow based on a recently-published manually-curated cell marker database [1], and incorporates a suite of visualization tools that can be overlaid on a t-SNE representation. The software is freely available as a python package at https://github.com/sdomanskyi/DigitalCellSorterConclusionsThis methodology assures that extensive marker to cell type matching information is taken into account in a systematic way when assigning cell clusters to cell types. Moreover, the method allows for a high throughput processing of multiple scRNA-seq datasets, since it does not involve an expert curator, and it can be applied recursively to obtain cell sub-types. The software is designed to allow the user to substitute the marker to cell type matching information and apply the methodology to different cellular environments.

scholarly journals BERMUDA: A novel deep transfer learning method for single-cell RNA sequencing batch correction reveals hidden high-resolution cellular subtypes

2019 ◽  
Author(s):  
Tongxin Wang ◽  
Travis S Johnson ◽  
Wei Shao ◽  
Zixiao Lu ◽  
Bryan R Helm ◽  
...  
Keyword(s):  
Single Cell ◽  
Rna Sequencing ◽  
Transfer Learning ◽  
Cell Types ◽  
Batch Effect ◽  
Batch Effects ◽  
Combine Data ◽  
Batch Correction ◽  
Single Cell Rna Sequencing ◽  
Bona Fide

AbstractTo fully utilize the power of single-cell RNA sequencing (scRNA-seq) technologies for cell lineation and identifyingbona fidetranscriptional signals, it is necessary to combine data from multiple experiments. We presentBERMUDA(Batch-Effect ReMoval Using Deep Autoencoders) — a novel transfer-learning-based method for batch-effect correction in scRNA-seq data.BERMUDAeffectively combines different batches of scRNA-seq data with vastly different cell population compositions and amplifies biological signals by transferring information among batches. We demonstrate thatBERMUDAoutperforms existing methods for removing batch effects and distinguishing cell types in multiple simulated and real scRNA-seq datasets.

scholarly journals IMMU-27. SINGLE CELL RNA-SEQUENCING IDENTIFIES NOVEL BONE MARROW DERIVED MYELOID CELLS IN GLIOBLASTOMA ASSOCIATED WITH TUMOR AGGRESSION

2020 ◽  
Vol 22 (Supplement_2) ◽  
pp. ii110-ii110
Author(s):  
Christina Jackson ◽  
Christopher Cherry ◽  
Sadhana Bom ◽  
Hao Zhang ◽  
John Choi ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Single Cell ◽  
Tumor Cells ◽  
Rna Sequencing ◽  
Metabolic Pathways ◽  
Myeloid Cells ◽  
Tumor Grade ◽  
Sequencing Data ◽  
Single Cell Rna Sequencing ◽  
Two Populations

Abstract BACKGROUND Glioma associated myeloid cells (GAMs) can be induced to adopt an immunosuppressive phenotype that can lead to inhibition of anti-tumor responses in glioblastoma (GBM). Understanding the composition and phenotypes of GAMs is essential to modulating the myeloid compartment as a therapeutic adjunct to improve anti-tumor immune response. METHODS We performed single-cell RNA-sequencing (sc-RNAseq) of 435,400 myeloid and tumor cells to identify transcriptomic and phenotypic differences in GAMs across glioma grades. We further correlated the heterogeneity of the GAM landscape with tumor cell transcriptomics to investigate interactions between GAMs and tumor cells. RESULTS sc-RNAseq revealed a diverse landscape of myeloid-lineage cells in gliomas with an increase in preponderance of bone marrow derived myeloid cells (BMDMs) with increasing tumor grade. We identified two populations of BMDMs unique to GBMs; Mac-1and Mac-2. Mac-1 demonstrates upregulation of immature myeloid gene signature and altered metabolic pathways. Mac-2 is characterized by expression of scavenger receptor MARCO. Pseudotime and RNA velocity analysis revealed the ability of Mac-1 to transition and differentiate to Mac-2 and other GAM subtypes. We further found that the presence of these two populations of BMDMs are associated with the presence of tumor cells with stem cell and mesenchymal features. Bulk RNA-sequencing data demonstrates that gene signatures of these populations are associated with worse survival in GBM. CONCLUSION We used sc-RNAseq to identify a novel population of immature BMDMs that is associated with higher glioma grades. This population exhibited altered metabolic pathways and stem-like potentials to differentiate into other GAM populations including GAMs with upregulation of immunosuppressive pathways. Our results elucidate unique interactions between BMDMs and GBM tumor cells that potentially drives GBM progression and the more aggressive mesenchymal subtype. Our discovery of these novel BMDMs have implications in new therapeutic targets in improving the efficacy of immune-based therapies in GBM.

scholarly journals Single-cell data clustering based on sparse optimization and low-rank matrix factorization

2021 ◽  
Author(s):  
Yinlei Hu ◽  
Bin Li ◽  
Falai Chen ◽  
Kun Qu
Keyword(s):  
Single Cell ◽  
Rna Sequencing ◽  
Matrix Factorization ◽  
Data Clustering ◽  
Cell Types ◽  
Low Rank ◽  
Sequencing Data ◽  
Rank Matrix ◽  
Single Cell Rna Sequencing ◽  
Low Rank Matrix

Abstract Unsupervised clustering is a fundamental step of single-cell RNA sequencing data analysis. This issue has inspired several clustering methods to classify cells in single-cell RNA sequencing data. However, accurate prediction of the cell clusters remains a substantial challenge. In this study, we propose a new algorithm for single-cell RNA sequencing data clustering based on Sparse Optimization and low-rank matrix factorization (scSO). We applied our scSO algorithm to analyze multiple benchmark datasets and showed that the cluster number predicted by scSO was close to the number of reference cell types and that most cells were correctly classified. Our scSO algorithm is available at https://github.com/QuKunLab/scSO. Overall, this study demonstrates a potent cell clustering approach that can help researchers distinguish cell types in single-cell RNA sequencing data.

scholarly journals Comprehensive network modeling from single cell RNA sequencing of human and mouse reveals well conserved transcription regulation of hematopoiesis

BMC Genomics ◽  
2020 ◽  
Vol 21 (S11) ◽  
Author(s):  
Shouguo Gao ◽  
Zhijie Wu ◽  
Xingmin Feng ◽  
Sachiko Kajigaya ◽  
Xujing Wang ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Transcription Factors ◽  
Single Cell ◽  
Rna Sequencing ◽  
Transcription Regulation ◽  
Regulatory Networks ◽  
Stem Cell Differentiation ◽  
Hematopoietic Stem ◽  
Single Cell Rna Sequencing ◽  
Human And Mouse

Abstract Background Presently, there is no comprehensive analysis of the transcription regulation network in hematopoiesis. Comparison of networks arising from gene co-expression across species can facilitate an understanding of the conservation of functional gene modules in hematopoiesis. Results We used single-cell RNA sequencing to profile bone marrow from human and mouse, and inferred transcription regulatory networks in each species in order to characterize transcriptional programs governing hematopoietic stem cell differentiation. We designed an algorithm for network reconstruction to conduct comparative transcriptomic analysis of hematopoietic gene co-expression and transcription regulation in human and mouse bone marrow cells. Co-expression network connectivity of hematopoiesis-related genes was found to be well conserved between mouse and human. The co-expression network showed “small-world” and “scale-free” architecture. The gene regulatory network formed a hierarchical structure, and hematopoiesis transcription factors localized to the hierarchy’s middle level. Conclusions Transcriptional regulatory networks are well conserved between human and mouse. The hierarchical organization of transcription factors may provide insights into hematopoietic cell lineage commitment, and to signal processing, cell survival and disease initiation.

scholarly journals Single cell RNA sequencing of the Strongylocentrotus purpuratus larva reveals the blueprint of major cell types and nervous system of a non-chordate deuterostome

eLife ◽  
2021 ◽  
Vol 10 ◽  
Author(s):  
Periklis Paganos ◽  
Danila Voronov ◽  
Jacob M Musser ◽  
Detlev Arendt ◽  
Maria Ina Arnone
Keyword(s):  
Single Cell ◽  
Rna Sequencing ◽  
Sea Urchin ◽  
Regulatory Networks ◽  
Sea Urchins ◽  
Neuronal Cell ◽  
Cell Types ◽  
Molecular Fingerprint ◽  
Larval Body ◽  
Single Cell Rna Sequencing

Identifying the molecular fingerprint of organismal cell types is key for understanding their function and evolution. Here, we use single cell RNA sequencing (scRNA-seq) to survey the cell types of the sea urchin early pluteus larva, representing an important developmental transition from non-feeding to feeding larva. We identify 21 distinct cell clusters, representing cells of the digestive, skeletal, immune, and nervous systems. Further subclustering of these reveal a highly detailed portrait of cell diversity across the larva, including the identification of neuronal cell types. We then validate important gene regulatory networks driving sea urchin development and reveal new domains of activity within the larval body. Focusing on neurons that co-express Pdx-1 and Brn1/2/4, we identify an unprecedented number of genes shared by this population of neurons in sea urchin and vertebrate endocrine pancreatic cells. Using differential expression results from Pdx-1 knockdown experiments, we show that Pdx1 is necessary for the acquisition of the neuronal identity of these cells. We hypothesize that a network similar to the one orchestrated by Pdx1 in the sea urchin neurons was active in an ancestral cell type and then inherited by neuronal and pancreatic developmental lineages in sea urchins and vertebrates.

Localization of migraine susceptibility genes in human brain by single-cell RNA sequencing

Cephalalgia ◽  
2018 ◽  
Vol 38 (13) ◽  
pp. 1976-1983 ◽  
Author(s):  
William Renthal
Keyword(s):  
Human Brain ◽  
Single Cell ◽  
Rna Sequencing ◽  
Expression Profiles ◽  
Cell Types ◽  
Susceptibility Genes ◽  
Brain Cell ◽  
Cell Type ◽  
Single Cell Rna Sequencing ◽  
Brain Cell Types

Background Migraine is a debilitating disorder characterized by severe headaches and associated neurological symptoms. A key challenge to understanding migraine has been the cellular complexity of the human brain and the multiple cell types implicated in its pathophysiology. The present study leverages recent advances in single-cell transcriptomics to localize the specific human brain cell types in which putative migraine susceptibility genes are expressed. Methods The cell-type specific expression of both familial and common migraine-associated genes was determined bioinformatically using data from 2,039 individual human brain cells across two published single-cell RNA sequencing datasets. Enrichment of migraine-associated genes was determined for each brain cell type. Results Analysis of single-brain cell RNA sequencing data from five major subtypes of cells in the human cortex (neurons, oligodendrocytes, astrocytes, microglia, and endothelial cells) indicates that over 40% of known migraine-associated genes are enriched in the expression profiles of a specific brain cell type. Further analysis of neuronal migraine-associated genes demonstrated that approximately 70% were significantly enriched in inhibitory neurons and 30% in excitatory neurons. Conclusions This study takes the next step in understanding the human brain cell types in which putative migraine susceptibility genes are expressed. Both familial and common migraine may arise from dysfunction of discrete cell types within the neurovascular unit, and localization of the affected cell type(s) in an individual patient may provide insight into to their susceptibility to migraine.

Identifying cell types to interpret scRNA-seq data: how, why and more possibilities

2020 ◽  
Vol 19 (4) ◽  
pp. 286-291 ◽  
Author(s):  
Ziwei Wang ◽  
Hui Ding ◽  
Quan Zou
Keyword(s):  
Single Cell ◽  
Rna Sequencing ◽  
Cell Types ◽  
Biological Phenomena ◽  
Single Cell Rna Sequencing ◽  
Numerous Data ◽  
Near Future

Abstract Single-cell RNA sequencing (scRNA-seq) has generated numerous data and renewed our understanding of biological phenomena at the cellular scale. Identification of cell types has been one of the most prevalent means for interpreting scRNA-seq data, based upon which connections are made between the transcriptome and phenotype. Herein, we attempt to review the methods and tools that dedicate to the task regarding their feature and usage and look at the possibilities for scRNA-seq development in the near future.

Sign in / Sign up

Export Citation Format

Share Document