Biology:Cellular deconvolution

From HandWiki
Short description: Set of computational techniques

Cellular deconvolution (also referred to as cell type composition or cell proportion estimation) refers to computational techniques aiming at estimating the proportions of different cell types in samples collected from a tissue.[1] For example, samples collected from the human brain are a mixture of various neuronal and glial cell types (e.g. microglia and astrocytes) in different proportions, where each cell type has a diverse gene expression profile.[2] Since most high-throughput technologies use bulk samples and measure the aggregated levels of molecular information (e.g. expression levels of genes) for all cells in a sample, the measured values would be an aggregate of the values pertaining to the expression landscape of different cell types.[3] Therefore, many downstream analyses such as differential gene expression might be confounded by the variations in cell type proportions when using the output of high-throughput technologies applied to bulk samples.[1] The development of statistical methods to identify cell type proportions in large-scale bulk samples is an important step for better understanding of the relationship between cell type composition and diseases.[4]

Cellular deconvolution algorithms have been applied to a variety of samples collected from saliva,[5] buccal,[5] cervical,[5] PBMC,[6] brain,[2] kidney,[1] and pancreatic cells,[1] and many studies have shown that estimating and incorporating the proportions of cell types into various analyses improves the interpretability of high-throughput omics data and reduces the confounding effects of cellular heterogeneity, also known as tissue heterogeneity, in functional analysis of omics data.[7][8]

Cellular Deconvolution Pipeline. All methods require the gene expression or DNA methylation profiles of each subject in the study using high-throughput technologies and bulk samples.

Mathematical Formulation

Most cellular deconvolution algorithms consider an input data in a form of a matrix [math]\displaystyle{ X_{m\times n} }[/math], which represents some molecular information (e.g. gene expression data or DNA methylation data) measured over a group of [math]\displaystyle{ n }[/math] samples and [math]\displaystyle{ m }[/math] marks (e.g. genes or CpG sites). The goal of the algorithm is to use these data and return an output matrix [math]\displaystyle{ W_{k\times n} }[/math], representing the proportions of [math]\displaystyle{ k }[/math] distinct cell types in each of the [math]\displaystyle{ n }[/math] samples. Some methods limit the sum of each column of [math]\displaystyle{ W }[/math] matrix less than or equal to one, so that the proportions of cell types sum up to the overall number of cells in the sample (less than one when there are some unknown cell types in the samples).[9] Moreover, it is assumed that the values of [math]\displaystyle{ W }[/math] matrix are non-negative as they pertain to proportions of cell types.[9]

Current strategies

There are two broad categories of methods aiming at estimating the proportion of cell types in samples using some type of omics data (bulk gene expression or DNA methylation data). These approaches are labeled as reference-based (also called supervised) and reference-free (also called unsupervised) methods[10][11]

Reference-based methods

Reference-based methods require an a priori defined reference matrix consisting of the expected value (also called profile or signature) of gene expression (or DNA methylation) for a group of genes (or CpG sites) known to have a differential expression (or methylation)

Reference-based methods and reference-free methods for cellular deconvolution. Reference-based approaches aim at estimating the contribution of each signature profile to the overall level of signal while reference-free methods need to estimate both latent cell type signatures and the contribution of each signature.[12]

across the cell types.[10] A reference matrix can be represented by a matrix [math]\displaystyle{ H_{m\times k} }[/math], representing the expected value for [math]\displaystyle{ m }[/math] markers (genes or CpG sites) for each of [math]\displaystyle{ k }[/math] cell types known to be presented in the samples. These references can be derived by exploring external single-cell epigenomics or transcriptomics datasets generated for a group of samples similar (e.g. in terms of biological condition, sex and age) to the samples for which the deconvolution method will be applied. These methods use statistical approaches such as non-negative or constrained linear regression methods to dissect the contribution of each cell type to the aggregated bulk signals of genes or CpG sites.[13] Constrained regression is the basis for many of reference-free cellular deconvolution methods existing in the literature, aiming at estimating the cell proportion values ([math]\displaystyle{ W_{k\times n} }[/math]) that maximizes the similarity between [math]\displaystyle{ HW^T }[/math] and [math]\displaystyle{ X }[/math].[13] The performance of reference-based methods depends critically on the quality of the reference profiles.[14]

Construction of reference profiles

There are a variety of approaches for isolating different cell types to measure their gene expression or DNA methylation levels to be used as references in the deconvolution algorithms. Earlier methods used cell sorting methods such as FACS (fluorescence-activated cell sorting) based on the flow cytometry technique, which separates the populations of cells belonging to different cell types based on their cell sizes, morphologies (shape), and surface protein expressions.[15][16][17] With the advance in single-cell technologies, newer approaches started to incorporate references for cell-types measured on a single-cell resolution obtained for a subset of subjects in the study or external subjects from a similar biological condition.[18][1][19]

Reference-free methods

Reference-free methods do not need the reference profiles of cell-type specific genes (or CpGs), although they might still require the identity (name) of cell-type-specific genes (or CpGs).[20] These methods might be considered as a modification of reference-based methods where both [math]\displaystyle{ H }[/math] and [math]\displaystyle{ W }[/math] are unknown, and the goal is to jointly estimate both matrices so that the similarity between [math]\displaystyle{ HW^T }[/math] and [math]\displaystyle{ X }[/math] is maximized. Many of the reference-free methods are based on mathematical framework of non-negative matrix factorization,[21][22][10] which imposes a non-negativity constraint on the elements of [math]\displaystyle{ H }[/math] and [math]\displaystyle{ W }[/math]. Additional constraints such as the assumption of orthogonality between the columns of [math]\displaystyle{ H }[/math] might be incorporated to improve the interpretability of results and prevent overfitting.

Some recent cellular deconvolution methods selected based on citations and publishing year.
Title Category Input Data Type Year
CIBERSORT[23] Robust enumeration of cell subsets from tissue expression profiles Reference based Gene expression 2018
CDSeq[24] A complete deconvolution method for dissecting tissue heterogeneity Reference free Gene expression 2019
FARDEEP[25] Fast and robust deconvolution of expression profiles Reference based Gene expression 2019
UNDO[26] Unsupervised deconvolution of tumor-stromal mixed expressions Reference free Gene expression 2015
dtangle[27] Accurate and robust cell type deconvolution Reference based Gene expression 2019
EPIC[28] Estimating the proportions of different cell types from bulk gene expression data Reference based Gene expression 2017
BSEQ-sc[29] Deconvolution of bulk sequencing experiments using single cell data Reference based Gene expression 2016
MuSiC[18] Cell-type Identification by estimating relative subsets of RNA transcripts Reference based Gene expression 2019
SCDC[30] Bulk gene expression deconvolution by multiple single-Cell RNA sequencing references Reference based Gene expression 2020
DWLS[31] Gene expression deconvolution using dampened weighted least squares Reference based Gene expression 2019
deconvSeq[32] Deconvolution of cell mixture distribution in sequencing data Reference based Gene expression 2019
Bisque[19] Decomposition of bulk expression with single-cell sequencing Reference based Gene expression 2020
TOAST[33] Tools for the analysis of heterogeneous tissues Reference free DNA methylation 2019
Houseman[9] Reference-free deconvolution of DNA methylation data and mediation by cell composition effects Reference based DNA methylation 2016
methylCC[34] Technology-independent estimation of cell type composition using differentially methylated regions Reference based DNA methylation 2019
BayesCCE[35] Bayesian framework for estimating cell-type composition from DNA methylation Reference free DNA methylation 2018

Advantages and limitations

Advantages

In silico cell-type level resolution

The advance of single-cell technologies enables the profiling of each individual cell in a sample, which help elucidate the issue of cellular heterogeneity by measuring the proportions of different cells in samples. Even though the quality of single cell profiling technologies has been on the rise in recent years, these technologies are still costly, limiting their applications in large populations of samples.[36] Single cell technologies such as single cell transcriptomic methods also tend to have higher error rates due to factors such as high dropout events.[37][38] Cellular deconvolution methods provide a robust and cost-effective in silico alternatives for understanding the samples on a cell-type level resolution, by relying on single cell information of only a small subset of cells in the sample, the reference profiles generated by external sources, or even no reference profile at all.[39]

(Re)analysis of old data

There are large amounts of old bulk data from studies concerning various diseases and biological conditions. These datasets could be considered important resources in studying of rare disease, long follow-up studies or samples and tissues that are difficult to extract. Since the biological samples for many of these studies are not available or accessible anymore, reprofiling the data using single cell technologies might not be within the realm of possibilities for many studies. Invention of more advanced cellular deconvolution methods gives the opportunity to researchers to come back to old omics studies, reanalyze their datasets, and scrutinize their findings.[39]

Limitations

Reliability of reference

Reference-based approaches rely on the availability of accurate references to estimate cell proportions. The discrepancy between the biology of the samples underlying the references and the samples for which the cell proportions are being estimated could introduce bias in estimated cell proportions.[40] Studies have shown that using references obtained from samples with different phenotypes such as age, gender, and disease status than the population of interest reduces the performance of reference-based methods to levels lower than their reference-free counterparts.[10][40]

Lack of reference for rare, unknown, or uncharacterized cell types

Reference-based approaches assume the existence of prior knowledge on the types of cells existing in a sample. Therefore, these methods may fail to perform accurately when the data includes rare or otherwise unknown cell types with no references incorporated in the algorithm.[7] For example, cancer tumors consist of heterogeneous mixtures of various healthy cells of different types such as immune cells and cells related to affected tissues in addition to tumor cells.[25] Although it might be possible to provide references for the immune cells, we do not usually have access to references or signatures for cancer cells due to the unique patterns of mutations and distributions of molecular information in each individual.[10] These situations have been addressed in some studies under the label of deconvolution methods with partial reference availability.[41]

Applications

Relationship between cell proportions and phenotypes

Confounding effect of cell proportions can leads to false associations between cortical gene expression and Alzheimer's disease clinical pathology.[2]

Studies have shown that the proportions of different cell types might show correlations with various phenotypes such as different diseases. For example, the proportions of Parathyroid oxyphil cells in the samples collected from the parathyroid gland for groups of patients show a significant correlation with the presence of clinical characteristics of chronic kidney disease (CKD).[42] Another study applying the cellular deconvolution algorithms to gene expression data of Alzheimer's patients find that patients with lower proportions of neuronal cells in the samples collected from their cerebral cortex are more likely to show the clinical characteristics of dementia.[43] Cellular deconvolution algorithms could enable researchers to investigate the interactions between cell proportions and various diseases or biological phenotypes.

Dissecting the confounding effects of cell proportions in EWAS and TWAS studies

Epigenome-wide association study (EWAS) and transcriptome-wide association studies (TWAS) aim at finding the molecular markers such as genes or methylation CpG sites that show significant correlations between their expression or methylation levels and a biological phenotype of interest such as a disease. Since the proportions of cell types in samples vary and might show significant correlations with the disease or phenotype of interest, these correlations may confound the functional relationships between genes or CpG sites and the disease or phenotypes under study.[44] For example, studies aimed at finding genes involved in Alzheimer's disease may end up selecting genes that are exclusively expressed in neurons and therefore have lower expression levels in Alzheimer's patients due to compositional changes of cell types during neurodegeneration.[45] Such genes are not actionable targets for the treatment of Alzheimer's since they are not causally involved in the biological mechanism underlying Alzheimer's disease, but are only brought up by the confounding effects of cell types.

References

  1. ↑ 1.0 1.1 1.2 1.3 1.4 "Benchmarking of cell type deconvolution pipelines for transcriptomics data". Nature Communications 11 (1): 5650. November 2020. doi:10.1038/s41467-020-19015-1. PMID 33159064. Bibcode: 2020NatCo..11.5650A. 
  2. ↑ 2.0 2.1 2.2 "Deconvolving the contributions of cell-type heterogeneity on cortical gene expression". PLOS Computational Biology 16 (8): e1008120. August 2020. doi:10.1371/journal.pcbi.1008120. PMID 32804935. Bibcode: 2020PLSCB..16E8120P. 
  3. ↑ "Cell population-specific expression analysis of human cerebellum". BMC Genomics 13 (1): 610. November 2012. doi:10.1186/1471-2164-13-610. PMID 23145530. 
  4. ↑ "Computational deconvolution of transcriptomics data from mixed cell populations". Bioinformatics 34 (11): 1969–1979. June 2018. doi:10.1093/bioinformatics/bty019. PMID 29351586. 
  5. ↑ 5.0 5.1 5.2 "A novel cell-type deconvolution algorithm reveals substantial contamination by immune cells in saliva, buccal and cervix". Epigenomics 10 (7): 925–940. July 2018. doi:10.2217/epi-2018-0037. PMID 29693419. 
  6. ↑ "Improved cell composition deconvolution method of bulk gene expression profiles to quantify subsets of immune cells". BMC Medical Genomics 12 (Suppl 8): 169. December 2019. doi:10.1186/s12920-019-0613-5. PMID 31856824. 
  7. ↑ 7.0 7.1 "Cellular deconvolution of GTEx tissues powers discovery of disease and cell-type associated regulatory variants". Nature Communications 11 (1): 955. February 2020. doi:10.1038/s41467-020-14561-0. PMID 32075962. Bibcode: 2020NatCo..11..955D. 
  8. ↑ "EPISCORE: cell type deconvolution of bulk tissue DNA methylomes from single-cell RNA-Seq data". Genome Biology 21 (1): 221. September 2020. doi:10.1186/s13059-020-02126-9. PMID 32883324. 
  9. ↑ 9.0 9.1 9.2 "Reference-free deconvolution of DNA methylation data and mediation by cell composition effects". BMC Bioinformatics 17 (1): 259. June 2016. doi:10.1186/s12859-016-1140-4. PMID 27358049. 
  10. ↑ 10.0 10.1 10.2 10.3 10.4 "Cell-type deconvolution in epigenome-wide association studies: a review and recommendations". Epigenomics 9 (5): 757–768. May 2017. doi:10.2217/epi-2016-0153. PMID 28517979. 
  11. ↑ "An Efficient and Flexible Method for Deconvoluting Bulk RNA-Seq Data with Single-Cell RNA-Seq Data". Cells 8 (10): 1161. September 2019. doi:10.3390/cells8101161. PMID 31569701. 
  12. ↑ "TOAST: improving reference-free cell composition estimation by cross-cell type differential analysis". Genome Biology 20 (1): 190. September 2019. doi:10.1186/s13059-019-1778-0. PMID 31484546. 
  13. ↑ 13.0 13.1 "Cell-type deconvolution from DNA methylation: a review of recent applications". Human Molecular Genetics 26 (R2): R216–R224. October 2017. doi:10.1093/hmg/ddx275. PMID 28977446. 
  14. ↑ Sturm, Gregor; Finotello, Francesca; Petitprez, Florent; Zhang, Jitao David; Baumbach, Jan; Fridman, Wolf H; List, Markus; Aneichyk, Tatsiana (15 July 2019). "Comprehensive evaluation of transcriptome-based cell-type quantification methods for immuno-oncology". Bioinformatics 35 (14): i436–i445. doi:10.1093/bioinformatics/btz363. PMID 31510660. 
  15. ↑ "Coral cell separation and isolation by fluorescence-activated cell sorting (FACS)". BMC Cell Biology 18 (1): 30. August 2017. doi:10.1186/s12860-017-0146-8. PMID 28851289. 
  16. ↑ "Differential DNA methylation in purified human blood cells: implications for cell lineage and studies on disease susceptibility". PLOS ONE 7 (7): e41361. 2012-07-25. doi:10.1371/journal.pone.0041361. PMID 22848472. Bibcode: 2012PLoSO...741361R. 
  17. ↑ "Improving cell mixture deconvolution by identifying optimal DNA methylation libraries (IDOL)". BMC Bioinformatics 17 (1): 120. March 2016. doi:10.1186/s12859-016-0943-7. PMID 26956433. 
  18. ↑ 18.0 18.1 "Bulk tissue cell type deconvolution with multi-subject single-cell expression reference". Nature Communications 10 (1): 380. January 2019. doi:10.1038/s41467-018-08023-x. PMID 30670690. Bibcode: 2019NatCo..10..380W. 
  19. ↑ 19.0 19.1 "Accurate estimation of cell composition in bulk expression through robust integration of single-cell information". Nature Communications 11 (1): 1971. April 2020. doi:10.1038/s41467-020-15816-6. PMID 32332754. Bibcode: 2020NatCo..11.1971J. 
  20. ↑ "NITUMID: Nonnegative matrix factorization-based Immune-TUmor MIcroenvironment Deconvolution". Bioinformatics 36 (5): 1344–1350. March 2020. doi:10.1093/bioinformatics/btz748. PMID 31593244. 
  21. ↑ "Biomarker discovery in heterogeneous tissue samples -taking the in-silico deconfounding approach". BMC Bioinformatics 11 (1): 27. January 2010. doi:10.1186/1471-2105-11-27. PMID 20070912. 
  22. ↑ "Virtual microdissection identifies distinct tumor- and stroma-specific subtypes of pancreatic ductal adenocarcinoma". Nature Genetics 47 (10): 1168–78. October 2015. doi:10.1038/ng.3398. PMID 26343385. 
  23. ↑ "Determining cell type abundance and expression from bulk tissues with digital cytometry". Nature Biotechnology 37 (7): 773–782. July 2019. doi:10.1038/s41587-019-0114-2. PMID 31061481. 
  24. ↑ "CDSeq: A novel complete deconvolution method for dissecting heterogeneous samples using gene expression data". PLOS Computational Biology 15 (12): e1007510. December 2019. doi:10.1371/journal.pcbi.1007510. PMID 31790389. Bibcode: 2019PLSCB..15E7510K. 
  25. ↑ 25.0 25.1 "Fast and robust deconvolution of tumor infiltrating lymphocyte from expression profiles using least trimmed squares". PLOS Computational Biology 15 (5): e1006976. May 2019. doi:10.1371/journal.pcbi.1006976. PMID 31059559. Bibcode: 2019PLSCB..15E6976H. 
  26. ↑ UNDO. Bioconductor. 2017. doi:10.18129/b9.bioc.undo. https://bioconductor.org/packages/UNDO. Retrieved 2021-02-23. 
  27. ↑ "dtangle: accurate and robust cell type deconvolution". Bioinformatics 35 (12): 2093–2099. June 2019. doi:10.1093/bioinformatics/bty926. PMID 30407492. 
  28. ↑ "Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data". eLife 6: e26476. November 2017. doi:10.7554/eLife.26476. PMID 29130882. 
  29. ↑ "A Single-Cell Transcriptomic Map of the Human and Mouse Pancreas Reveals Inter- and Intra-cell Population Structure". Cell Systems 3 (4): 346–360.e4. October 2016. doi:10.1016/j.cels.2016.08.011. PMID 27667365. 
  30. ↑ "SCDC: bulk gene expression deconvolution by multiple single-cell RNA sequencing references". Briefings in Bioinformatics 22 (1): 416–427. January 2021. doi:10.1093/bib/bbz166. PMID 31925417. 
  31. ↑ "Accurate estimation of cell-type composition from gene expression data". Nature Communications 10 (1): 2975. July 2019. doi:10.1038/s41467-019-10802-z. PMID 31278265. Bibcode: 2019NatCo..10.2975T. 
  32. ↑ "deconvSeq: deconvolution of cell mixture distribution in sequencing data". Bioinformatics 35 (24): 5095–5102. December 2019. doi:10.1093/bioinformatics/btz444. PMID 31147676. 
  33. ↑ "Dissecting differential signals in high-throughput data from complex tissues". Bioinformatics 35 (20): 3898–3905. October 2019. doi:10.1093/bioinformatics/btz196. PMID 30903684. 
  34. ↑ "methylCC: technology-independent estimation of cell type composition using differentially methylated regions". Genome Biology 20 (1): 261. November 2019. doi:10.1186/s13059-019-1827-8. PMID 31783894. 
  35. ↑ "BayesCCE: a Bayesian framework for estimating cell-type composition from DNA methylation without the need for methylation reference". Genome Biology 19 (1): 141. September 2018. doi:10.1186/s13059-018-1513-2. PMID 30241486. 
  36. ↑ "Bulk tissue cell type deconvolution with multi-subject single-cell expression reference". Nature Communications 10 (1): 380. January 2019. doi:10.1038/s41467-018-08023-x. PMID 30670690. Bibcode: 2019NatCo..10..380W. 
  37. ↑ "scDoc: correcting drop-out events in single-cell RNA-seq data". Bioinformatics 36 (15): 4233–4239. August 2020. doi:10.1093/bioinformatics/btaa283. PMID 32365169. 
  38. ↑ "Systematic comparison of high-throughput single-cell RNA-seq methods for immune cell profiling". BMC Genomics 22 (1): 66. January 2021. doi:10.1186/s12864-020-07358-4. PMID 33472597. 
  39. ↑ 39.0 39.1 "Bayesian estimation of cell-type-specific gene expression per bulk sample with prior derived from single-cell data.". bioRxiv. January 2020. doi:10.1101/2020.08.05.238949. 
  40. ↑ 40.0 40.1 "Systematic evaluation and validation of reference and library selection methods for deconvolution of cord blood DNA methylation data". Clinical Epigenetics 11 (1): 125. August 2019. doi:10.1186/s13148-019-0717-y. PMID 31455416. 
  41. ↑ "Deconvolution of heterogeneous tumor samples using partial reference signals". PLOS Computational Biology 16 (11): e1008452. November 2020. doi:10.1371/journal.pcbi.1008452. PMID 33253170. Bibcode: 2020PLSCB..16E8452Q. 
  42. ↑ "Relationship between parathyroid oxyphil cell proportion and clinical characteristics of patients with chronic kidney disease". International Urology and Nephrology 52 (1): 155–159. January 2020. doi:10.1007/s11255-019-02330-y. PMID 31686279. 
  43. ↑ "Cell number changes in Alzheimer's disease relate to dementia, not to plaques and tangles". Brain 136 (Pt 12): 3738–52. December 2013. doi:10.1093/brain/awt273. PMID 24136825. 
  44. ↑ "Cell-Type Heterogeneity in Adipose Tissue Is Associated with Complex Traits and Reveals Disease-Relevant Cell-Specific eQTLs". American Journal of Human Genetics 104 (6): 1013–1024. June 2019. doi:10.1016/j.ajhg.2019.03.025. PMID 31130283. 
  45. ↑ "Single-cell transcriptomic analysis of Alzheimer's disease". Nature 570 (7761): 332–337. June 2019. doi:10.1038/s41586-019-1195-2. PMID 31042697. Bibcode: 2019Natur.570..332M. 



Retrieved from "https://handwiki.org/wiki/index.php?title=Biology:Cellular_deconvolution&oldid=3517146"