genomics - single cell genomics II

advantages of single-cell multi-omics

-more subpopulations may be identified, as different technologies may pick up different types of variations

-link the genome and its epigenetic regulation to gene and protein expression at the single-cell level

CRISPR + single cell genomics

use crispr/cas9 technology to knockout/edit genes or non coding regions (enhancers)

-infect pools of cells with viral constructs containing guide rnas that target specific areas of the genome

-use scRNA-seq to profile the transcriptome of each cell + the specific guide RNAs that were transduced

-linking gene expression changes with the factor being manipulated

V(D)J recombination and pared TCR sequences

VDJ recombination: somatic recombination of variable, diversity, and joining gene sequences in B/T cells

-partition cells into GEMs. all cDNA generated in a single GEM share a common 10x barcode

-perform enrichment PCR targeting the 5’ end to the C-region

-followed by enzymatic fragmentation results in a pool of molecules originating from the same transcript

-the molecules carry the same 10x carvode and UMi sequences, but w different insert lengths, resulting in different sequence start points

-the diversity of start points gives complete coverage of the targeted portion of each VDJ transcript, which is ~650 bp

single cell immune profiling

simultaneous analysis of following libraries at single cell resolution for the same set of cells

-VDJ transcripts and clonotypes for B or T cells

-5’ single cell gene expression

-cell surface proteins (antibody capture)

-barcode enabled antigen mapping (antigen capture)

-crispr guide capture

spatial genomics

-the position of any given cell, relative to its neighbors and non-cellular structures, can provide helpful information for defining cellular phenotype, cell state, and ultimately cell and tissue function

-provide spatially resolved, high-dimensional assessment of transcripts, proteins, or metabolites

microarray-based spatial transcriptomics benefits

tissue placed on a barcoded glass slide

-high spatial resolution

-little specialized equipment

-unbiased

microarray-based spatial transcriptomics limitations

low capture efficiency

low resolution compared to FISH

microfluids based spatial transcriptomics benefits

co-mapping capability

high resolution

high genes/pixel

microfluids based spatial transcriptomics limitations

near single-cell resolution

tissue size is limited

FISH spatial transcriptomics benefits

fluorescence in situ hybridization

-high capture efficiency

-subcellular resolution

FISH limitations

readout limited to targeted genes

marker gene count

transcript length

LCM-based spatial transcriptomics benefits

Laser capture microdissection

preservation of tissue morphology

quick

high resolution

LCM-based spatial transcriptomics limitations

costly

sample quality is a limitation

Visium

-enables whole transcriptome analysis of entire sections

-protein co-detection

-not a single-cell resolution: each barcoded spot captures the transcripts from 1-10 cells

-relatively low sensitivity

-high cost

-labor intensive process

visium gene and protein expression co-detection

-enables protein and whole transcriptome RNA mapped together in a single experiment from a single tissue section

-pre validated, 35 plex antibody panel optimized for use on human FFPE tissues

-visium hd slide uses a panel of predesigned probes, capture area w continuous lawn of oligos

stereo-seq

-poly dT based capture method

-enables unbiased transcriptome exploration

-ideal for studying diverse biological systems

-resolution of 0.5 um, require higher sequencing depth

imaging based technologies : merscope

merfish: multiplexed error-robust fluorescence in situ hybridization

-fluorescent labeled probes + combinatorial barcode + sequential hybridization and imaging

sensitivity

detection efficiency when compared to the total number of expressed transcripts

specificity

the fraction of reported transcripts that correspond to a true transcript within the biological sample

-errors from stray autofluorescence in the tissue

-incomplete probe binding

-molecular crowding

information density

the amount of information that can be measured within a given tissue volume or area

effective multiplexing capacity

the number of RNAs a spatial genomics technology can detect in a single experiment

true multiplexing capacity determined by

-targeted multiplexing capacity

-the noise floor (ambient background signal)

data analysis for sequencing based technologies

-do not capture single cell resolution, the gene expression profile of spots reflects cell-type composition rather than distinct cell types

-estimates the cell-type composition per spot based on the gene expression profile of the cell populations in a single cell resolved reference

data analysis for imaging based technologies

-predefined set of transcripts

-cell segmentation

-similar to scRNA-seq data analysis

-imputation of the whole transcriptome (measured in standard scRNA-seq) in a spatially resolved manner

advantages of imaging based methods over sequencing based

-better concordance with CODEX

-higher accuracy cell segmentation using DAPI images

-higher proportion of transcripts confined within the segmented cells

-better clustering quality

-higher spatial resolution

spatial proteomics - imaging mass cytometry (IMC)

-use of metal tagged antibodies (40+ proteins)

-single cell resolution

-no autofluorescence

-protein and rna co-detection

-cons: small imaging area

spatial proteomics - phenocycler (codex)

-integrating automated fluidics and iterative imaging

-use barcoded antibody (100+ proteins)

-single-cell resolution

-imaging 1 million cells in 10 minutes

-whole slide imaging

spatial proteomics data analysis

-image pre processing: image registration and correction of experimental and imaging artifacts, including channel crosstalk, background, noise etc

-cell segmentation: facilitated by recent advances in AI algorithms, primarily trained on large manually curated datasets

-feature quantification: the mean intensity of each protein, cell size, circularity, location

-cell classification: normalization, filtering for lineage-based proteins, gating, clustering, supervised machine learning or probabilistic modeling

-cell phenotyping: annotate cells based on marker proteins

-spatial analysis: cell to cell interactions, define microenvironments, segment microanatomical structures and stratify patients for clinical insights

spatial metabolomics

-enabled localizing metabolites, lipids, and drugs in tissue sections

-has reached 5-10 um resolution

-single-cell metabolomics just started