scholarly journals Automated identification of cell-type–specific genes and alternative promoters

2021 ◽  
Author(s):  
Mickaël Mendez ◽  
Jayson Harshbarger ◽  
Michael M. Hoffman

Background: Identifying key transcriptional features, such as genes or transcripts, involved in cellular differentiation remains a challenging problem. Current methods for identifying key transcriptional features predominantly rely on pairwise comparisons among different cell types. These methods also identify long lists of differentially expressed transcriptional features. Combining the results from many such pairwise comparisons to find the transcriptional features specific only to one cell type is not straightforward. Thus, one must have a principled method for amalgamating pairwise cell type comparisons that makes full use of prior knowledge about the developmental relationships between cell types. Method: We developed Cell Lineage Analysis (CLA), a computational method which identifies transcriptional features with expression patterns that discriminate cell types, incorporating Cell Ontology knowledge on the relationship between different cell types. CLA uses random forest classification with a stratified bootstrap to increase the accuracy of binary classifiers when each cell type have a different number of samples. Regularized random forest results in a classifier that selects few but important transcriptional features. For each cell type pair, CLA runs multiple instances of regularized random forest and reports the transcriptional features consistently selected. CLA not only discriminates individual cell types but can also discriminate lineages of cell types related in the developmental hierarchy. Results: We applied CLA to Functional Annotation of the Mammalian Genome 5 (FANTOM5) data and identified discriminative transcription factor and long non-coding RNA (lncRNA) genes for 71 human cell types. With capped analysis of gene expression (CAGE) data, CLA identified individual cell-type–specific alternative promoters for cell surface markers. Compared to random forest with a standard bootstrap approach, CLA's stratified bootstrap approach improved the accuracy of gene expression classification models for more than 95% of 2060 cell type pairs examined. Applied on 10X Genomics single-cell RNA-seq data for CD14+ monocytes and FCGR3A+ monocytes, CLA selected only 13 discriminative genes. These genes included the top 9 out of 370 significantly differentially expressed genes obtained from conventional differential expression analysis methods. Discussion: Our CLA method combines tools to simplify the interpretation of transcriptome datasets from many cell types. It automates the identification of the most differentially expressed genes for each cell type pairs CLA's lineage score allows easy identification of the best transcriptional markers for each cell type and lineage in both bulk and single-cell transcriptomic data. Availability: CLA is available at https://cla.hoffmanlab.org. We deposited the version of the CLA source with which we ran our experiments at https://doi.org/10.5281/zenodo.3630670. We deposited other analysis code and results at https://doi.org/10.5281/zenodo.5735636.

2018 ◽  
Author(s):  
Xiangyu Luo ◽  
Can Yang ◽  
Yingying Wei

In epigenome-wide association studies, the measured signals for each sample are a mixture of methylation profiles from different cell types. The current approaches to the association detection only claim whether a cytosine-phosphate-guanine (CpG) site is associated with the phenotype or not, but they cannot determine the cell type in which the risk-CpG site is affected by the phenotype. Here, we propose a solid statistical method, HIgh REsolution (HIRE), which not only substantially improves the power of association detection at the aggregated level as compared to the existing methods but also enables the detection of risk-CpG sites for individual cell types.


2020 ◽  
Author(s):  
Yupeng Wang ◽  
Rosario B. Jaime-Lara ◽  
Abhrarup Roy ◽  
Ying Sun ◽  
Xinyue Liu ◽  
...  

AbstractWe propose SeqEnhDL, a deep learning framework for classifying cell type-specific enhancers based on sequence features. DNA sequences of “strong enhancer” chromatin states in nine cell types from the ENCODE project were retrieved to build and test enhancer classifiers. For any DNA sequence, sequential k-mer (k=5, 7, 9 and 11) fold changes relative to randomly selected non-coding sequences were used as features for deep learning models. Three deep learning models were implemented, including multi-layer perceptron (MLP), Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). All models in SeqEnhDL outperform state-of-the-art enhancer classifiers including gkm-SVM and DanQ, with regard to distinguishing cell type-specific enhancers from randomly selected non-coding sequences. Moreover, SeqEnhDL is able to directly discriminate enhancers from different cell types, which has not been achieved by other enhancer classifiers. Our analysis suggests that both enhancers and their tissue-specificity can be accurately identified according to their sequence features. SeqEnhDL is publicly available at https://github.com/wyp1125/SeqEnhDL.


1985 ◽  
Vol 101 (4) ◽  
pp. 1442-1454 ◽  
Author(s):  
P Cowin ◽  
H P Kapprell ◽  
W W Franke

Desmosomal plaque proteins have been identified in immunoblotting and immunolocalization experiments on a wide range of cell types from several species, using a panel of monoclonal murine antibodies to desmoplakins I and II and a guinea pig antiserum to desmosomal band 5 protein. Specifically, we have taken advantage of the fact that certain antibodies react with both desmoplakins I and II, whereas others react only with desmoplakin I, indicating that desmoplakin I contains unique regions not present on the closely related desmoplakin II. While some of these antibodies recognize epitopes conserved between chick and man, others display a narrow species specificity. The results show that proteins whose size, charge, and biochemical behavior are very similar to those of desmoplakin I and band 5 protein of cow snout epidermis are present in all desmosomes examined. These include examples of simple and pseudostratified epithelia and myocardial tissue, in addition to those of stratified epithelia. In contrast, in immunoblotting experiments, we have detected desmoplakin II only among cells of stratified and pseudostratified epithelial tissues. This suggests that the desmosomal plaque structure varies in its complement of polypeptides in a cell-type specific manner. We conclude that the obligatory desmosomal plaque proteins, desmoplakin I and band 5 protein, are expressed in a coordinate fashion but independently from other differentiation programs of expression such as those specific for either epithelial or cardiac cells.


2019 ◽  
Author(s):  
Qi Song ◽  
Jiyoung Lee ◽  
Shamima Akter ◽  
Ruth Grene ◽  
Song Li

AbstractRecent advances in genomic technologies have generated large-scale protein-DNA interaction data and open chromatic regions for multiple plant species. To predict condition specific gene regulatory networks using these data, we developed the Condition Specific Regulatory network inference engine (ConSReg), which combines heterogeneous genomic data using sparse linear model followed by feature selection and stability selection to select key regulatory genes. Using Arabidopsis as a model system, we constructed maps of gene regulation under more than 50 experimental conditions including abiotic stresses, cell type-specific expression, and stress responses in individual cell types. Our results show that ConSReg accurately predicted gene expressions (average auROC of 0.84) across multiple testing datasets. We found that, (1) including open chromatin information from ATAC-seq data significantly improves the performance of ConSReg across all tested datasets; (2) choice of negative training samples and length of promoter regions are two key factors that affect model performance. We applied ConSReg to Arabidopsis single cell RNA-seq data of two root cell types (endodermis and cortex) and identified five regulators in two root cell types. Four out of the five regulators have additional experimental evidence to support their roles in regulating gene expression in Arabidopsis roots. By comparing regulatory maps in abiotic stress responses and cell type-specific experiments, we revealed that transcription factors that regulate tissue levels abiotic stresses tend to also regulate stress responses in individual cell types in plants.


2021 ◽  
Author(s):  
Maize C Cao ◽  
Emma L Scotter

TDP-43 proteinopathy is the major pathological hallmark of amyotrophic lateral sclerosis (ALS) and tau-negative frontotemporal dementia (FTD). Mounting evidence implicates a loss of normal TDP-43 function in neurodegeneration, either resultant from or independent of TDP-43 aggregation. TDP-43 knockdown is therefore a common paradigm for modelling ALS and FTD. However, because TDP-43 can interact directly with thousands of mRNA targets and regulate the function of other RNA binding proteins, the phenotype of TDP-43 depletion is likely to differ depending on the proteomic and transcriptomic profile of the model cell type. Here, we conducted a meta-analysis of publicly available RNA-sequencing datasets that utilized TDP-43 knockdown to model ALS or FTD, and validated these against RNA-sequencing data from TDP-43-immunonegative neuronal nuclei from ALS/FTD brain. We present these analyses as easy-to-use interactive graphical databases. Of 9 TDP-43-knockdown datasets identified, 4 showed significant depletion of TARDBP (human HeLa and SH-SY5Y cell lines, induced human motor neurons, and mouse striatal tissue). There was little overlap in differentially expressed genes between TDP-43-knockdown model cell types, but PFKP, RANBP1, KIAA1324, ELAVL3, and STMN2 were among the common TDP-43 targets. Of these, only STMN2 was validated as a differentially expressed gene in TDP-43-immunonegative neuronal nuclei in ALS/FTD brain. Similarly, there were few genes that showed common patterns of differential exon usage between cell types and which validated in TDP-43-immunonegative neurons, but these included well-known targets POLDIP3, RANBP1, STMN2, and UNC13A, and novel targets EXD3, CEP290, KPNA4, and MMAB. Enrichment analysis showed that TDP-43 knockdown in different cell types affected a unique range of biological pathways. Together, these data identify novel TDP-43 targets, validate known TDP-43 targets, and show that TDP-43 plays both conserved and cell-type-specific roles in the regulation of gene expression and splicing. Identification of cell-type-specific TDP-43 targets will enable sensitive mapping of cell-autonomous TDP-43 dysfunction beyond just neurons, while shared TDP-43 targets are likely to have therapeutic value across myriad cell types.


2020 ◽  
Author(s):  
Jiaxin Fan ◽  
Xuran Wang ◽  
Rui Xiao ◽  
Mingyao Li

AbstractAllelic expression imbalance (AEI), quantified by the relative expression of two alleles of a gene in a diploid organism, can help explain phenotypic variations among individuals. Traditional methods detect AEI using bulk RNA sequencing (RNA-seq) data, a data type that averages out cell-to-cell heterogeneity in gene expression across cell types. Since the patterns of AEI may vary across different cell types, it is desirable to study AEI in a cell-type-specific manner. Although this can be achieved by single-cell RNA sequencing (scRNA-seq), it requires full-length transcript to be sequenced in single cells of a large number of individuals, which are still cost prohibitive to generate. To overcome this limitation and utilize the vast amount of existing disease relevant bulk tissue RNA-seq data, we developed BSCET, which enables the characterization of cell-type-specific AEI in bulk RNA-seq data by integrating cell type composition information inferred from a small set of scRNA-seq samples, possibly obtained from an external dataset. By modeling covariate effect, BSCET can also detect genes whose cell-type-specific AEI are associated with clinical factors. Through extensive benchmark evaluations, we show that BSCET correctly detected genes with cell-type-specific AEI and differential AEI between healthy and diseased samples using bulk RNA-seq data. BSCET also uncovered cell-type-specific AEIs that were missed in bulk data analysis when the directions of AEI are opposite in different cell types. We further applied BSCET to two pancreatic islet bulk RNA-seq datasets, and detected genes showing cell-type-specific AEI that are related to the progression of type 2 diabetes. Since bulk RNA-seq data are easily accessible, BSCET provided a convenient tool to integrate information from scRNA-seq data to gain insight on AEI with cell type resolution. Results from such analysis will advance our understanding of cell type contributions in human diseases.Author SummaryDetection of allelic expression imbalance (AEI), a phenomenon where the two alleles of a gene differ in their expression magnitude, is a key step towards the understanding of phenotypic variations among individuals. Existing methods detect AEI use bulk RNA sequencing (RNA-seq) data and ignore AEI variations among different cell types. Although single-cell RNA sequencing (scRNA-seq) has enabled the characterization of cell-to-cell heterogeneity in gene expression, the high costs have limited its application in AEI analysis. To overcome this limitation, we developed BSCET to characterize cell-type-specific AEI using the widely available bulk RNA-seq data by integrating cell-type composition information inferred from scRNA-seq samples. Since the degree of AEI may vary with disease phenotypes, we further extended BSCET to detect genes whose cell-type-specific AEIs are associated with clinical factors. Through extensive benchmark evaluations and analyses of two pancreatic islet bulk RNA-seq datasets, we demonstrated BSCET’s ability to refine bulk-level AEI to cell-type resolution, and to identify genes whose cell-type-specific AEIs are associated with the progression of type 2 diabetes. With the vast amount of easily accessible bulk RNA-seq data, we believe BSCET will be a valuable tool for elucidating cell type contributions in human diseases.


PLoS Genetics ◽  
2021 ◽  
Vol 17 (3) ◽  
pp. e1009080
Author(s):  
Jiaxin Fan ◽  
Xuran Wang ◽  
Rui Xiao ◽  
Mingyao Li

Allelic expression imbalance (AEI), quantified by the relative expression of two alleles of a gene in a diploid organism, can help explain phenotypic variations among individuals. Traditional methods detect AEI using bulk RNA sequencing (RNA-seq) data, a data type that averages out cell-to-cell heterogeneity in gene expression across cell types. Since the patterns of AEI may vary across different cell types, it is desirable to study AEI in a cell-type-specific manner. Although this can be achieved by single-cell RNA sequencing (scRNA-seq), it requires full-length transcript to be sequenced in single cells of a large number of individuals, which are still cost prohibitive to generate. To overcome this limitation and utilize the vast amount of existing disease relevant bulk tissue RNA-seq data, we developed BSCET, which enables the characterization of cell-type-specific AEI in bulk RNA-seq data by integrating cell type composition information inferred from a small set of scRNA-seq samples, possibly obtained from an external dataset. By modeling covariate effect, BSCET can also detect genes whose cell-type-specific AEI are associated with clinical factors. Through extensive benchmark evaluations, we show that BSCET correctly detected genes with cell-type-specific AEI and differential AEI between healthy and diseased samples using bulk RNA-seq data. BSCET also uncovered cell-type-specific AEIs that were missed in bulk data analysis when the directions of AEI are opposite in different cell types. We further applied BSCET to two pancreatic islet bulk RNA-seq datasets, and detected genes showing cell-type-specific AEI that are related to the progression of type 2 diabetes. Since bulk RNA-seq data are easily accessible, BSCET provided a convenient tool to integrate information from scRNA-seq data to gain insight on AEI with cell type resolution. Results from such analysis will advance our understanding of cell type contributions in human diseases.


2021 ◽  
Vol 15 ◽  
Author(s):  
Mingchao Li ◽  
Qing Min ◽  
Matthew C. Banton ◽  
Xinpeng Dun

Advances in single-cell RNA sequencing technologies and bioinformatics methods allow for both the identification of cell types in a complex tissue and the large-scale gene expression profiling of various cell types in a mixture. In this report, we analyzed a single-cell RNA sequencing (scRNA-seq) dataset for the intact adult mouse sciatic nerve and examined cell-type specific transcription factor expression and activity during peripheral nerve homeostasis. In total, we identified 238 transcription factors expressed in nine different cell types of intact mouse sciatic nerve. Vascular smooth muscle cells have the lowest number of transcription factors expressed with 17 transcription factors identified. Myelinating Schwann cells (mSCs) have the highest number of transcription factors expressed, with 61 transcription factors identified. We created a cell-type specific expression map for the identified 238 transcription factors. Our results not only provide valuable information about the expression pattern of transcription factors in different cell types of adult peripheral nerves but also facilitate future studies to understand the function of key transcription factors in the peripheral nerve homeostasis and disease.


eLife ◽  
2015 ◽  
Vol 4 ◽  
Author(s):  
Alex de Mendoza ◽  
Hiroshi Suga ◽  
Jon Permanyer ◽  
Manuel Irimia ◽  
Iñaki Ruiz-Trillo

Cell-type specification through differential genome regulation is a hallmark of complex multicellularity. However, it remains unclear how this process evolved during the transition from unicellular to multicellular organisms. To address this question, we investigated transcriptional dynamics in the ichthyosporean Creolimax fragrantissima, a relative of animals that undergoes coenocytic development. We find that Creolimax utilizes dynamic regulation of alternative splicing, long inter-genic non-coding RNAs and co-regulated gene modules associated with animal multicellularity in a cell-type specific manner. Moreover, our study suggests that the different cell types of the three closest animal relatives (ichthyosporeans, filastereans and choanoflagellates) are the product of lineage-specific innovations. Additionally, a proteomic survey of the secretome reveals adaptations to a fungal-like lifestyle. In summary, the diversity of cell types among protistan relatives of animals and their complex genome regulation demonstrates that the last unicellular ancestor of animals was already capable of elaborate specification of cell types.


2020 ◽  
Vol 49 (D1) ◽  
pp. D1413-D1419 ◽  
Author(s):  
Tianyi Zhao ◽  
Shuxuan Lyu ◽  
Guilin Lu ◽  
Liran Juan ◽  
Xi Zeng ◽  
...  

Abstract SC2disease (http://easybioai.com/sc2disease/) is a manually curated database that aims to provide a comprehensive and accurate resource of gene expression profiles in various cell types for different diseases. With the development of single-cell RNA sequencing (scRNA-seq) technologies, uncovering cellular heterogeneity of different tissues for different diseases has become feasible by profiling transcriptomes across cell types at the cellular level. In particular, comparing gene expression profiles between different cell types and identifying cell-type-specific genes in various diseases offers new possibilities to address biological and medical questions. However, systematic, hierarchical and vast databases of gene expression profiles in human diseases at the cellular level are lacking. Thus, we reviewed the literature prior to March 2020 for studies which used scRNA-seq to study diseases with human samples, and developed the SC2disease database to summarize all the data by different diseases, tissues and cell types. SC2disease documents 946 481 entries, corresponding to 341 cell types, 29 tissues and 25 diseases. Each entry in the SC2disease database contains comparisons of differentially expressed genes between different cell types, tissues and disease-related health status. Furthermore, we reanalyzed gene expression matrix by unified pipeline to improve the comparability between different studies. For each disease, we also compare cell-type-specific genes with the corresponding genes of lead single nucleotide polymorphisms (SNPs) identified in genome-wide association studies (GWAS) to implicate cell type specificity of the traits.


Sign in / Sign up

Export Citation Format

Share Document