Dataset Information

SIDEseq: A Cell Similarity Measure Defined by Shared Identified Differentially Expressed Genes for Single-Cell RNA sequencing Data.

ABSTRACT: One goal of single-cell RNA sequencing (scRNA seq) is to expose possible heterogeneity within cell populations due to meaningful, biological variation. Examining cell-to-cell heterogeneity, and further, identifying subpopulations of cells based on scRNA seq data has been of common interest in life science research. A key component to successfully identifying cell subpopulations (or clustering cells) is the (dis)similarity measure used to group the cells. In this paper, we introduce a novel measure, named SIDEseq, to assess cell-to-cell similarity using scRNA seq data. SIDEseq first identifies a list of putative differentially expressed (DE) genes for each pair of cells. SIDEseq then integrates the information from all the DE gene lists (corresponding to all pairs of cells) to build a similarity measure between two cells. SIDEseq can be implemented in any clustering algorithm that requires a (dis)similarity matrix. This new measure incorporates information from all cells when evaluating the similarity between any two cells, a characteristic not commonly found in existing (dis)similarity measures. This property is advantageous for two reasons: (a) borrowing information from cells of different subpopulations allows for the investigation of pairwise cell relationships from a global perspective and (b) information from other cells of the same subpopulation could help to ensure a robust relationship assessment. We applied SIDEseq to a newly generated human ovarian cancer scRNA seq dataset, a public human embryo scRNA seq dataset, and several simulated datasets. The clustering results suggest that the SIDEseq measure is capable of uncovering important relationships between cells, and outperforms or at least does as well as several popular (dis)similarity measures when used on these datasets.

SUBMITTER: Schiffman C

PROVIDER: S-EPMC6377168 | biostudies-literature | 2017 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

SIDEseq: A Cell Similarity Measure Defined by Shared Identified Differentially Expressed Genes for Single-Cell RNA sequencing Data.

Schiffman Courtney C Lin Christina C Shi Funan F Chen Luonan L Sohn Lydia L Huang Haiyan H

Statistics in biosciences 20170517 1

One goal of single-cell RNA sequencing (scRNA seq) is to expose possible heterogeneity within cell populations due to meaningful, biological variation. Examining cell-to-cell heterogeneity, and further, identifying subpopulations of cells based on scRNA seq data has been of common interest in life science research. A key component to successfully identifying cell subpopulations (or clustering cells) is the (dis)similarity measure used to group the cells. In this paper, we introduce a novel measu ...[more]

PMID: 30774736

Similar Datasets

Project description:In the search for a therapeutic schedule for spinal cord injury, it is necessary to understand key genes and their corresponding regulatory networks involved in the spinal cord injury process. However, ad hoc selection and analysis of one or two genes cannot fully reveal the complex molecular biological mechanisms of spinal cord injury. The emergence of second-generation sequencing technology (RNA sequencing) has provided a better method. In this study, RNA sequencing technology was used to analyze differentially expressed genes at different time points after spinal cord injury in rat models established by contusion of the eighth thoracic segment. The numbers of genes that changed significantly were 944, 1362 and 1421 at 1, 4 and 7 days after spinal cord injury respectively. After gene ontology analysis and temporal expression analysis of the differentially expressed genes, C5ar1, Socs3 and CCL6 genes were then selected and identified by real-time polymerase chain reaction and western blot assay. The mRNA expression trends of C5ar1, Socs3 and CCL6 genes were consistent with the RNA sequencing results. Further verification and analysis of C5ar1 indicate that the level of protein expression of C5ar1 was consistent with its nucleic acid level after spinal cord injury. C5ar1 was mainly expressed in neurons and astrocytes. Finally, the gene Itgb2, which may be related to C5ar1, was found by Chilibot database and literature search. Immunofluorescence histochemical results showed that the expression of Itgb2 was highly consistent with that of C5ar1. Itgb2 was expressed in astrocytes. RNA sequencing technology can screen differentially expressed genes at different time points after spinal cord injury. Through analysis and verification, genes strongly associated with spinal cord injury can be screened. This can provide experimental data for further determining the molecular mechanism of spinal cord injury, and also provide possible targets for the treatment of spinal cord injury. This study was approved ethically by the Laboratory Animal Ethics Committee of Jiangsu Province, China (approval No. 2018-0306-001) on March 6, 2018.

Project description:As a type of osteoporosis (OP), male idiopathic OP (MIO) is a bone disorder that occurs in young males and is a public health problem worldwide. However, the detailed pathogenesis of MIO remains to be elucidated. In the present study, blood samples of patients with MIO, senile OP, postmenopausal OP and normal controls (NCs) were obtained for RNA sequencing. Compared with the NC group, differentially expressed genes (DEGs) in the three types of OP were identified. DEGs that were common among the three types of OP and the DEGs that were unique to patients with MIO were determined. Gene ontology enrichment analysis and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses were conducted. MIO?specific and OP?specific protein?protein interaction (PPI) networks were constructed. Compared with NCs, a total of 519, 368 and 1,472 DEGs were identified in samples from MIO, senile OP and postmenopausal OP, respectively. Tetraspanin 5 (TSPAN5) and ??synuclein (SNCA) were unique DEGs in MIO that were not identified in the other two types of OP compared with NCs. Furthermore, the expression of carbonic anhydrase 1 (CA1) and S100 calcium?binding protein P (S100P) in MIO was significantly different compared with senile OP, postmenopausal OP and NC samples. 'MAPK signaling pathway', 'type I diabetes mellitus' and 'hematopoietic cell lineage' were among significantly enriched pathways of DEGs in MIO. SNCA and CDC?like kinase 1 were the hub genes in the MIO?specific PPI network. In conclusion, the mitogen?activated protein kinase signaling and type I diabetes mellitus pathways may be involved in bone formation; SNCA and TSPAN5 may be associated with bone resorption. These two pathways and two genes may serve a role in MIO. CA1 and S100P may regulate the process of MIO by modulation of calcification and dysregulation of calcium binding. These findings may have provided an experimental basis for elucidating the underlying mechanisms and developing potential diagnostic biomarkers of MIO.

Project description:Analyses of Expressed Sequence Tags (ESTs) databases suggest that most human genes have multiple alternative splice variants. The alternative splicing of pre-mRNA is tightly regulated during development and in different tissue types. Changes in splicing patterns have been described in disease states. Recently, we used whole-transcriptome shotgun pryrosequencing to characterize 4 malignant pleural mesothelioma (MPM) tumors, 1 lung adenocarcinoma and 1 normal lung. We hypothesized that alternative splicing profiles might be detected in the sequencing data for the expressed genes in these samples.We developed a software pipeline to map the transcriptome read sequences of the 4 MPM samples and 1 normal lung sample onto known exon junction sequences in the comprehensive AceView database of expressed sequences and to count how many reads map to each junction. 13,274,187 transcriptome reads generated by the Roche/454 sequencing platform for 5 samples were compared with 151,486 exon junctions from the AceView database. The exon junction expression index (EJEI) was calculated for each exon junction in each sample to measure the differential expression of alternative splicing events. Top ten exon junctions with the largest EJEI difference between the 4 mesothelioma and the normal lung sample were then examined for differential expression using Quantitative Real Time PCR (qRT-PCR) in the 5 sequenced samples. Two of the differentially expressed exon junctions (ACTG2.aAug05 and CDK4.aAug05) were further examined with qRT-PCR in additional 18 MPM and 18 normal lung specimens.We found 70,953 exon junctions covered by at least one sequence read in at least one of the 5 samples. All 10 identified most differentially expressed exon junctions were validated as present by RT-PCR, and 8 were differentially expressed exactly as predicted by the sequence analysis. The differential expression of the AceView exon junctions for the ACTG2 and CDK4 genes were also observed to be statistically significant in an additional 18 MPM and 18 normal lung samples examined using qRT-PCR. The differential expression of these two junctions was shown to successfully classify these mesothelioma and normal lung specimens with high sensitivity (89% and 78%, respectively).Whole-transcriptome shotgun sequencing, combined with a downstream bioinformatics pipeline, provides powerful tools for the identification of differentially expressed exon junctions resulting from alternative splice variants. The alternatively spliced genes discovered in the study could serve as useful diagnostic markers as well as potential therapeutic targets for MPM.

Dataset Information

SIDEseq: A Cell Similarity Measure Defined by Shared Identified Differentially Expressed Genes for Single-Cell RNA sequencing Data.

Publications

SIDEseq: A Cell Similarity Measure Defined by Shared Identified Differentially Expressed Genes for Single-Cell RNA sequencing Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets