Tutorial: guidelines for annotating single-cell transcriptomic maps using automated and manual methods

Clarke, Zoe Alixandra; Andrews, Tallulah; Atif, Jawairia; Pouyabahar, Delaram; Innes, Brendan T.; MacParland, Sonya A.; Bader, Gary D.

doi:10.31219/osf.io/ndafw

Cited by 31 publications

(45 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…14b,c ).This is relevant as automated cell type annotation methods often produce unreliable results in challenging cell populations, such as doublets, rare cell types, or transitional cell states, and require manual assessment. It is usually done by plotting marker expression, which requires corrected data for comparability 70 , while expiMap scores are directly comparable. Similarly, expiMap scores can resolve coarsely-annotated cell types; in our case the B cells that we annotated under the joint term of immune cells ( Fig.…”

Section: Resultsmentioning

confidence: 99%

“…We used expiMap to integrate three non-T2D pancreatic datasets differing in multiple biological factors, including sex, age, and stress status (Fltp_P16, NOD, and spikein_drug, from here on named as the reference datasets) using PanglaoDB marker sets and Reactome pathways. PanglaoDB marker sets were used to enable cell type identification, as previously proposed for scRNA-seq annotation 70,71 . As an alternative, we used Reactome pathways to enable the identification of molecular processes 72 differentially active across biological conditions.…”

Section: Expimap Delineates Pancreatic Cell and Subtypes After Diseas...mentioning

confidence: 99%

See 1 more Smart Citation

Biologically informed deep learning to infer gene program activity in single cells

Lotfollahi

Rybakov

Hrovatin

et al. 2022

Preprint

View full text Add to dashboard Cite

The increasing availability of large-scale single-cell datasets has enabled the detailed description of cell states across multiple biological conditions and perturbations. In parallel, recent advances in unsupervised machine learning, particularly in transfer learning, have enabled fast and scalable mapping of these new single-cell datasets onto reference atlases. The resulting large-scale machine learning models however often have millions of parameters, rendering interpretation of the newly mapped datasets challenging. Here, we propose expiMap, a deep learning model that enables interpretable reference mapping using biologically understandable entities, such as curated sets of genes and gene programs. The key concept is the substitution of the uninterpretable nodes in an autoencoder's bottleneck by labeled nodes mapping to interpretable lists of genes, such as gene ontologies, biological pathways, or curated gene sets, for which activities are learned as constraints during reconstruction. This is enabled by the incorporation of predefined gene programs into the reference model, and at the same time allowing the model to learn de novo new programs and refine existing programs durin reference mapping. We show that the model retains similar integration performance as existing methods while providing a biologically interpretable framework for understanding cellular behavior. We demonstrate the capabilities of expiMap by applying it to 15 datasets encompassing five different tissues and species. The interpretable nature of the mapping revealed unreported associations between interferon signaling via the RIG-I/MDA5 and GPCRs pathways, with differential behavior in CD8+ T cells and CD14+ monocytes in severe COVID-19, as well as the role of annexins in the cellular communications between lymphoid and myeloid compartments for explaining patient response to the applied drugs. Finally, expiMap enabled the direct comparison of a diverse set of pancreatic beta cells from multiple studies where we observed a strong, previously unreported correlation between the unfolded protein response and asparagine N-linked glycosylation. Altogether, expiMap enables the interpretable mapping of single cell transcriptome data sets across cohorts, disease states and other perturbations.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Expimap Delineates Pancreatic Cell and Subtypes After Diseas...mentioning

confidence: 99%

Biologically informed deep learning to infer gene program activity in single cells

Lotfollahi

Rybakov

Hrovatin

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The majority of markers in each cluster were specific except for a few that overlapped between clusters. The overlapping markers may be due to proliferating precursor cells that share transcriptional profiles, even though they are destined to form different cell types (Clarke et al, 2021).…”

Section: Discussionmentioning

confidence: 99%

Single cell analysis of gene expression in the substantia nigra pars compacta of a pesticide-induced mouse model of Parkinson’s disease

Ah¹,

Lee²,

Smith³

2022

Preprint

View full text Add to dashboard Cite

Exposure to pesticides in humans increases the risk of Parkinson's disease (PD), but the mechanisms remain poorly understood. To elucidate these pathways, we dosed C57BL/6J mice with a combination of the pesticides maneb and paraquat (MNPQ). Behavioral analysis revealed motor deficits consistent with PD. Single cell RNA sequencing of substantia nigra pars compacta revealed both cell-type specific genes and genes expressed differentially between pesticide and control, including Fam241b, Emx2os, Bivm, Gm1439, Prdm15 and Rai2. Neurons had the largest number of significant differentially expressed genes, but comparable numbers were found in astrocytes and less so in oligodendrocytes. In addition, network analysis revealed enrichment in functions related to the extracellular matrix. These findings emphasize the importance of support cells in pesticide-induced PD and refocus our attention away from neurons as the sole agent of this disorder.

show abstract

“…After clustering, assigning a biological annotation to each cluster is the basis of the subsequent analysis. Generally, the workflow for annotating cells in scRNA‐seq data includes three main steps 60 : automatic annotation, manual annotation and validation with wet experiments. Firstly, major automated annotation tools utilize a pre‐defined set of marker genes that are specifically expressed in a known cell type to label clusters by matching their gene expression patterns to known cell types.…”

Section: Streamline Scrna‐seq Data Analysismentioning

confidence: 99%

Single‐cell RNA sequencing technologies and applications: A brief overview

Jovic

Liang

Zeng

et al. 2022

Clinical & Translational Med

512

204

View full text Add to dashboard Cite

Single‐cell RNA sequencing (scRNA‐seq) technology has become the state‐of‐the‐art approach for unravelling the heterogeneity and complexity of RNA transcripts within individual cells, as well as revealing the composition of different cell types and functions within highly organized tissues/organs/organisms. Since its first discovery in 2009, studies based on scRNA‐seq provide massive information across different fields making exciting new discoveries in better understanding the composition and interaction of cells within humans, model animals and plants. In this review, we provide a concise overview about the scRNA‐seq technology, experimental and computational procedures for transforming the biological and molecular processes into computational and statistical data. We also provide an explanation of the key technological steps in implementing the technology. We highlight a few examples on how scRNA‐seq can provide unique information for better understanding health and diseases. One important application of the scRNA‐seq technology is to build a better and high‐resolution catalogue of cells in all living organism, commonly known as atlas, which is key resource to better understand and provide a solution in treating diseases. While great promises have been demonstrated with the technology in all areas, we further highlight a few remaining challenges to be overcome and its great potentials in transforming current protocols in disease diagnosis and treatment.

show abstract

Tutorial: guidelines for annotating single-cell transcriptomic maps using automated and manual methods

Cited by 31 publications

References 0 publications

Biologically informed deep learning to infer gene program activity in single cells

Biologically informed deep learning to infer gene program activity in single cells

Single cell analysis of gene expression in the substantia nigra pars compacta of a pesticide-induced mouse model of Parkinson’s disease

Single‐cell RNA sequencing technologies and applications: A brief overview

Contact Info

Product

Resources

About