scholarly journals Interpretable Autoencoders Trained on Single Cell Sequencing Data Can Transfer Directly to Data from Unseen Tissues

Cells ◽  
2021 ◽  
Vol 11 (1) ◽  
pp. 85
Author(s):  
Julie Sparholt Walbech ◽  
Savvas Kinalis ◽  
Ole Winther ◽  
Finn Cilius Nielsen ◽  
Frederik Otzen Bagger
Keyword(s):  
Single Cell ◽  
Cell Types ◽  
Biological Pathways ◽  
Training Data ◽  
Sequencing Data ◽  
Data Simulation ◽  
Single Cell Sequencing ◽  
Saliency Maps ◽  
Unseen Data ◽  
Biological Concepts

Autoencoders have been used to model single-cell mRNA-sequencing data with the purpose of denoising, visualization, data simulation, and dimensionality reduction. We, and others, have shown that autoencoders can be explainable models and interpreted in terms of biology. Here, we show that such autoencoders can generalize to the extent that they can transfer directly without additional training. In practice, we can extract biological modules, denoise, and classify data correctly from an autoencoder that was trained on a different dataset and with different cells (a foreign model). We deconvoluted the biological signal encoded in the bottleneck layer of scRNA-models using saliency maps and mapped salient features to biological pathways. Biological concepts could be associated with specific nodes and interpreted in relation to biological pathways. Even in this unsupervised framework, with no prior information about cell types or labels, the specific biological pathways deduced from the model were in line with findings in previous research. It was hypothesized that autoencoders could learn and represent meaningful biology; here, we show with a systematic experiment that this is true and even transcends the training data. This means that carefully trained autoencoders can be used to assist the interpretation of new unseen data.

484 Bioturing browser: interactively explore public single cell sequencing data

2020 ◽  
Vol 8 (Suppl 3) ◽  
pp. A520-A520
Author(s):  
Son Pham ◽  
Tri Le ◽  
Tan Phan ◽  
Minh Pham ◽  
Huy Nguyen ◽  
...  
Keyword(s):  
Single Cell ◽  
Immune Cell ◽  
Expression Profiles ◽  
Meta Analysis ◽  
Cell Types ◽  
Sequencing Data ◽  
Single Cell Sequencing ◽  
Data Formats ◽  
Cancer Types ◽  
Cell Data

BackgroundSingle-cell sequencing technology has opened an unprecedented ability to interrogate cancer. It reveals significant insights into the intratumoral heterogeneity, metastasis, therapeutic resistance, which facilitates target discovery and validation in cancer treatment. With rapid advancements in throughput and strategies, a particular immuno-oncology study can produce multi-omics profiles for several thousands of individual cells. This overflow of single-cell data poses formidable challenges, including standardizing data formats across studies, performing reanalysis for individual datasets and meta-analysis.MethodsN/AResultsWe present BioTuring Browser, an interactive platform for accessing and reanalyzing published single-cell omics data. The platform is currently hosting a curated database of more than 10 million cells from 247 projects, covering more than 120 immune cell types and subtypes, and 15 different cancer types. All data are processed and annotated with standardized labels of cell types, diseases, therapeutic responses, etc. to be instantly accessed and explored in a uniform visualization and analytics interface. Based on this massive curated database, BioTuring Browser supports searching similar expression profiles, querying a target across datasets and automatic cell type annotation. The platform supports single-cell RNA-seq, CITE-seq and TCR-seq data. BioTuring Browser is now available for download at www.bioturing.com.ConclusionsN/A

scholarly journals Integrated analysis of bulk multi omic and single-cell sequencing data confirms the molecular origin of hemodynamic changes in Covid-19 infection explaining coagulopathy and higher geriatric mortality

2020 ◽  
Author(s):  
Shreya Johri ◽  
Deepali Jain ◽  
Ishaan Gupta
Keyword(s):  
Gene Expression ◽  
Single Cell ◽  
Older Patients ◽  
Cell Types ◽  
Integrated Analysis ◽  
Molecular Evidence ◽  
Sequencing Data ◽  
Phagocytic Cells ◽  
Hemodynamic Changes ◽  
Single Cell Sequencing

AbstractBesides severe respiratory distress, recent reports in Covid-19 patients have found a strong association between platelet counts and patient survival. Along with hemodynamic changes such as prolonged clotting time, high fibrin degradation products and D-dimers, increased levels of monocytes with disturbed morphology have also been identified. In this study, through an integrated analysis of bulk RNA-sequencing data from Covid-19 patients with data from single-cell sequencing studies on lung tissues, we found that most of the cell-types that contributed to the altered gene expression were of hematopoietic origin. We also found that differentially expressed genes in Covid-19 patients formed a significant pool of the expressing genes in phagocytic cells such as Monocytes and platelets. Interestingly, while we observed a general enrichment for Monocytes in Covid-19 patients, we found that the signal for FCGRA3+ Monocytes was depleted. Further, we found evidence that age-associated gene expression changes in Monocytes and platelets, associated with inflammation, mirror gene expression changes in Covid-19 patients suggesting that pro-inflammatory signalling during aging may worsen the infection in older patients. We identified more than 20 genes that change in the same direction between Covid-19 infection and aging cells that may act as potential therapeutic targets. Of particular interest were IL2RG, GNLY and GMZA expressed in platelets, which facilitates cytokine signalling in Monocytes through an interaction with platelets. To understand whether infection can directly manipulate the biology of Monocytes and platelets, we hypothesize that these non-ACE2 expressing cells may be infected by the virus through the phagocytic route. We observed that phagocytic cells such as Monocytes, T-cells, and platelets have a significantly higher expression of genes that are a part of the Covid-19 viral interactome. Hence these cell-types may have an active rather than a reactive role in viral pathogenesis to manifest clinical symptoms such as coagulopathy. Therefore, our results present molecular evidence for pursuing both anti-inflammatory and anticoagulation therapy for better patient management especially in older patients.

scholarly journals Learning for single-cell assignment

2020 ◽  
Vol 6 (44) ◽  
pp. eabd0855
Author(s):  
Bin Duan ◽  
Chenyu Zhu ◽  
Guohui Chuai ◽  
Chen Tang ◽  
Xiaohan Chen ◽  
...  
Keyword(s):  
Single Cell ◽  
Marker Gene ◽  
Cell Types ◽  
Sequencing Data ◽  
Single Cell Sequencing ◽  
Limited Effectiveness ◽  
Benchmark Datasets ◽  
Cell Assignment ◽  
Single Cell Type ◽  
Sequencing Data Analysis

Efficient single-cell assignment without prior marker gene annotations is essential for single-cell sequencing data analysis. Current methods, however, have limited effectiveness for distinct single-cell assignment. They failed to achieve a well-generalized performance in different tasks because of the inherent heterogeneity of different single-cell sequencing datasets and different single-cell types. Furthermore, current methods are inefficient to identify novel cell types that are absent in the reference datasets. To this end, we present scLearn, a learning-based framework that automatically infers quantitative measurement/similarity and threshold that can be used for different single-cell assignment tasks, achieving a well-generalized assignment performance on different single-cell types. We evaluated scLearn on a comprehensive set of publicly available benchmark datasets. We proved that scLearn outperformed the comparable existing methods for single-cell assignment from various aspects, demonstrating state-of-the-art effectiveness with a reliable and generalized single-cell type identification and categorizing ability.

scholarly journals Combination of single cell sequencing data and GWAS summary statistics reveals genetically-influenced liver cell types for primary biliary cholangitis

2021 ◽  
Author(s):  
Bingyu Xiang ◽  
Chunyu Deng ◽  
Jingjing Li ◽  
Shanshan Li ◽  
Huifang Zhang ◽  
...  
Keyword(s):  
T Cell ◽  
Single Cell ◽  
Liver Cell ◽  
Cell Types ◽  
Liver Cells ◽  
Primary Biliary Cholangitis ◽  
Summary Statistics ◽  
Sequencing Data ◽  
Immune Microenvironment ◽  
Single Cell Sequencing

Many genome-wide association studies (GWAS) have reported that numerous genetic loci were significantly associated with primary biliary cholangitis (PBC). However, the effects of genetic determinants on liver cells and its immune microenvironment for PBC remain unclear. We constructed a powerful computational framework to integrate a large-scale GWAS summary statistics (N = 13,239) with scRNA-seq data to uncover genetics-modulated liver cell subpopulations for PBC. We found that 29 genes including ORMDL3, GSNK2B, and DDAH2 were significantly associated with PBC susceptibility. Gene-property analysis revealed that four immune cell types including Cst3+ dendritic cell, Chil3+ macrophage, Trbc2+ T cell, and Gzma+ T cell were significantly enriched by PBC-risk genes. By combining GWAS summary statistics with scRNA-seq data, we identified that cholangiocytes exhibited a notable enrichment by PBC-related genetic association signals. The ORMDL3 gene showed the highest expression proportion in cholangiocytes than other liver cells (22.38%). Compared with ORMDL3+ cholangiocytes, we identified that ORMDL3- cholangiocytes predispose to play important immune-modulatory roles in the etiology of PBC. To the best of our knowledge, this is the first study to integrate human genetic information with single cell sequencing data for parsing genetics-influenced liver cells and its immune microenvironment for PBC risk.

scholarly journals Single-cell transcriptomic analysis of mIHC images via antigen mapping

2021 ◽  
Vol 7 (10) ◽  
pp. eabc5464
Author(s):  
Kiya W. Govek ◽  
Emma C. Troisi ◽  
Zhen Miao ◽  
Rachael G. Aubin ◽  
Steven Woodhouse ◽  
...  
Keyword(s):  
Single Cell ◽  
Spatial Patterns ◽  
Cell Types ◽  
Level Of Detail ◽  
Cell Populations ◽  
Sequencing Data ◽  
Spatially Resolved ◽  
Murine Spleen ◽  
Single Cell Rna Sequencing ◽  
Antibody Panel

Highly multiplexed immunohistochemistry (mIHC) enables the staining and quantification of dozens of antigens in a tissue section with single-cell resolution. However, annotating cell populations that differ little in the profiled antigens or for which the antibody panel does not include specific markers is challenging. To overcome this obstacle, we have developed an approach for enriching mIHC images with single-cell RNA sequencing data, building upon recent experimental procedures for augmenting single-cell transcriptomes with concurrent antigen measurements. Spatially-resolved Transcriptomics via Epitope Anchoring (STvEA) performs transcriptome-guided annotation of highly multiplexed cytometry datasets. It increases the level of detail in histological analyses by enabling the systematic annotation of nuanced cell populations, spatial patterns of transcription, and interactions between cell types. We demonstrate the utility of STvEA by uncovering the architecture of poorly characterized cell types in the murine spleen using published cytometry and mIHC data of this organ.

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 Effective clustering for single cell sequencing cancer data

2019 ◽  
Author(s):  
Simone Ciccolella ◽  
Murray Patterson ◽  
Paola Bonizzoni ◽  
Gianluca Della Vedova
Keyword(s):  
Single Cell ◽  
Categorical Data ◽  
Euclidean Distance ◽  
Missing Values ◽  
False Negative ◽  
Ground Truth ◽  
Sequencing Data ◽  
Large Space ◽  
Cancer Data ◽  
Single Cell Sequencing

AbstractBackgroundSingle cell sequencing (SCS) technologies provide a level of resolution that makes it indispensable for inferring from a sequenced tumor, evolutionary trees or phylogenies representing an accumulation of cancerous mutations. A drawback of SCS is elevated false negative and missing value rates, resulting in a large space of possible solutions, which in turn makes infeasible using some approaches and tools. While this has not inhibited the development of methods for inferring phylogenies from SCS data, the continuing increase in size and resolution of these data begin to put a strain on such methods.One possible solution is to reduce the size of an SCS instance — usually represented as a matrix of presence, absence and missing values of the mutations found in the different sequenced cells — and infer the tree from this reduced-size instance. Previous approaches have used k-means to this end, clustering groups of mutations and/or cells, and using these means as the reduced instance. Such an approach typically uses the Euclidean distance for computing means. However, since the values in these matrices are of a categorical nature (having the three categories: present, absent and missing), we explore techniques for clustering categorical data — commonly used in data mining and machine learning — to SCS data, with this goal in mind.ResultsIn this work, we present a new clustering procedure aimed at clustering categorical vector, or matrix data — here representing SCS instances, called celluloid. We demonstrate that celluloid clusters mutations with high precision: never pairing too many mutations that are unrelated in the ground truth, but also obtains accurate results in terms of the phylogeny inferred downstream from the reduced instance produced by this method.Finally, we demonstrate the usefulness of a clustering step by applying the entire pipeline (clustering + inference method) to a real dataset, showing a significant reduction in the runtime, raising considerably the upper bound on the size of SCS instances which can be solved in practice.AvailabilityOur approach, celluloid: clustering single cell sequencing data around centroids is available at https://github.com/AlgoLab/celluloid/ under an MIT license.

scholarly journals Identifying common and novel cell types in single-cell RNA-sequencing data using FR-Match

2021 ◽  
Author(s):  
Yun Zhang ◽  
Brian Aevermann ◽  
Rohan Gala ◽  
Richard H. Scheuermann
Keyword(s):  
Single Cell ◽  
Cell Types ◽  
Sample Type ◽  
Cell Type ◽  
Sequencing Data ◽  
Excellent Performance ◽  
Single Cell Rna Sequencing ◽  
Accurate Performance ◽  
Cross Platform ◽  
Tissue Region

Reference cell type atlases powered by single cell transcriptomic profiling technologies have become available to study cellular diversity at a granular level. We present FR-Match for matching query datasets to reference atlases with robust and accurate performance for identifying novel cell types and non-optimally clustered cell types in the query data. This approach shows excellent performance for cross-platform, cross-sample type, cross-tissue region, and cross-data modality cell type matching.

scholarly journals K-mer counting with low memory consumption enables fast clustering of single-cell sequencing data without read alignment

2019 ◽  
Author(s):  
Christina Huan Shi ◽  
Kevin Y. Yip
Keyword(s):  
Single Cell ◽  
State Of The Art ◽  
Rna Seq ◽  
Sequencing Data ◽  
Memory Consumption ◽  
Analysis Pipeline ◽  
Cell Clusters ◽  
Single Cell Sequencing ◽  
Sequencing Errors ◽  
Full Analysis

AbstractK-mer counting has many applications in sequencing data processing and analysis. However, sequencing errors can produce many false k-mers that substantially increase the memory requirement during counting. We propose a fast k-mer counting method, CQF-deNoise, which has a novel component for dynamically identifying and removing false k-mers while preserving counting accuracy. Compared with four state-of-the-art k-mer counting methods, CQF-deNoise consumed 49-76% less memory than the second best method, but still ran competitively fast. The k-mer counts from CQF-deNoise produced cell clusters from single-cell RNA-seq data highly consistent with CellRanger but required only 5% of the running time at the same memory consumption, suggesting that CQF-deNoise can be used for a preview of cell clusters for an early detection of potential data problems, before running a much more time-consuming full analysis pipeline.

Sign in / Sign up

Export Citation Format

Share Document